Learn AI Engineering! Learn it. Build it. Ship it for others.
This report summarizes scanner evidence for rohitg00/ai-engineering-from-scratch, a Python project with 66,834 GitHub stars. SynthScan examined 775,115 lines of code across 5366 source files, recording 6973 pattern matches distributed across 22 syntactic categories. The legacy adjusted pattern signal is 13.9 (Moderate pattern signal); it is not an authorship estimate.
Use the findings to ask whether maintainers can explain intent, verify dependencies, simplify unnecessary scaffolding, and focus review effort. Risk Payload v1 adds ownership, supply-chain, over-engineering, and review-effort vectors when available. These are prompts for human verification, not probabilities, security findings, or measured maintenance costs.
Use each vector to guide manual verification of the underlying findings.
Can maintainers explain intent, assumptions, and trade-offs? 4524 finding(s).
Are imports declared and package/API names verified? 1563 finding(s).
Are wrappers, guards, and repeated structures necessary? 3664 finding(s).
Which changes need focused human review? 6973 finding(s).
This chart tracks the legacy adjusted pattern score across scan runs. Changes can reflect source changes, scan coverage, or detector/scoring versions; they do not measure AI adoption or prove that a team refactored code. Compare runs only when their scoring versions are compatible. The dashed line tracks the raw pattern-hit count.
Groups findings by configured severity weight. Severity is not a confidence estimate, security rating, or proof that a package is missing. Verify imports against the project's manifests and inspect each finding in its local design context.
This horizontal bar chart shows where configured pattern scores accumulate across directories. Treat the bars as locations to inspect, not as evidence of authorship or code quality. Use the file-level findings to decide whether a directory needs a focused ownership, dependency, complexity, or review check.
The scanner identified 6973 distinct pattern matches across 22 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a detector match to inspect in context. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.
Reading the findings table: The Severity column indicates the configured finding weight (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — legacy comment-context matches receive a ×1.5 score weight. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier under the legacy scoring model.
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW⚡ | …/04-schedule-under-deadlines/tests/test_integration.py | 9 | def test_real_callback_reservation_and_release(self): | CODE |
| LOW⚡ | …/04-schedule-under-deadlines/tests/test_integration.py | 16 | def test_denied_job_never_invokes(self): | CODE |
| LOW⚡ | …/04-schedule-under-deadlines/tests/test_integration.py | 22 | def test_uncertain_receipt_retains_hold(self): | CODE |
| LOW⚡ | …/04-schedule-under-deadlines/tests/test_integration.py | 29 | def test_overrun_requires_reconciliation(self): | CODE |
| LOW⚡ | …/04-schedule-under-deadlines/tests/test_integration.py | 33 | def test_deadline_before_dispatch(self): | CODE |
| LOW | …04-schedule-under-deadlines/tests/test_event_status.py | 15 | def test_replay_reports_every_gate_through_status(self): | CODE |
| LOW | …04-schedule-under-deadlines/tests/test_event_status.py | 32 | def test_cli_modes_share_the_outcome_field(self): | CODE |
| LOW⚡ | …ge-desk/stages/03-draft-for-review/tests/test_stage.py | 30 | def test_unsent_email_round_trips(self): | CODE |
| LOW⚡ | …ge-desk/stages/03-draft-for-review/tests/test_stage.py | 38 | def test_reply_contains_source_and_edit_placeholder(self): | CODE |
| LOW⚡ | …ge-desk/stages/03-draft-for-review/tests/test_stage.py | 43 | def test_existing_reply_subject_not_duplicated(self): | CODE |
| LOW⚡ | …ge-desk/stages/03-draft-for-review/tests/test_stage.py | 48 | def test_wrong_identity_and_fabricated_span_rejected(self): | CODE |
| LOW⚡ | …ge-desk/stages/03-draft-for-review/tests/test_stage.py | 58 | def test_provider_request_and_response_contract(self): | CODE |
| LOW | …ge-desk/stages/03-draft-for-review/tests/test_stage.py | 83 | def test_provider_fabrication_and_endpoint_rejected(self): | CODE |
| LOW | …desk/stages/03-draft-for-review/tests/test_endpoint.py | 21 | def test_remote_http_and_malformed_authorities_fail_before_transport(self): | CODE |
| LOW | …desk/stages/03-draft-for-review/tests/test_endpoint.py | 51 | def test_https_and_explicit_loopback_http_are_supported(self): | CODE |
| LOW | …desk/stages/03-draft-for-review/tests/test_endpoint.py | 76 | def test_redirects_never_contact_another_endpoint(self): | CODE |
| LOW | …age-desk/stages/04-export-the-desk/tests/test_stage.py | 20 | def test_prioritized_queue_and_real_draft_files(self): | CODE |
| LOW | …age-desk/stages/04-export-the-desk/tests/test_stage.py | 38 | def test_html_escapes_subject_body_and_reason_evidence(self): | CODE |
| LOW⚡ | …age-desk/stages/04-export-the-desk/tests/test_stage.py | 55 | def test_subject_cannot_control_draft_path(self): | CODE |
| LOW⚡ | …age-desk/stages/04-export-the-desk/tests/test_stage.py | 61 | def test_json_preserves_source_and_method(self): | CODE |
| LOW⚡ | …age-desk/stages/04-export-the-desk/tests/test_stage.py | 71 | def test_empty_export_is_valid(self): | CODE |
| LOW⚡ | …age-desk/stages/04-export-the-desk/tests/test_stage.py | 77 | def test_identical_duplicates_write_one_entry(self): | CODE |
| LOW⚡ | …iage-desk/stages/01-parse-messages/tests/test_stage.py | 15 | def test_plain_headers_and_body(self): | CODE |
| LOW⚡ | …iage-desk/stages/01-parse-messages/tests/test_stage.py | 22 | def test_multipart_prefers_plain_and_ignores_attachment(self): | CODE |
| LOW⚡ | …iage-desk/stages/01-parse-messages/tests/test_stage.py | 32 | def test_html_only_requires_manual_review(self): | CODE |
| LOW⚡ | …iage-desk/stages/01-parse-messages/tests/test_stage.py | 40 | def test_missing_id_is_stable_and_body_sensitive(self): | CODE |
| LOW⚡ | …iage-desk/stages/01-parse-messages/tests/test_stage.py | 46 | def test_reference_ids_deduplicate(self): | CODE |
| LOW⚡ | …iage-desk/stages/01-parse-messages/tests/test_stage.py | 56 | def test_empty_missing_sender_and_oversize_rejected(self): | CODE |
| LOW⚡ | …ge-desk/stages/02-group-and-triage/tests/test_stage.py | 17 | def test_transitive_thread_and_unrelated_subject(self): | CODE |
| LOW⚡ | …ge-desk/stages/02-group-and-triage/tests/test_stage.py | 24 | def test_duplicate_and_conflicting_identity(self): | CODE |
| LOW⚡ | …ge-desk/stages/02-group-and-triage/tests/test_stage.py | 29 | def test_cycles_and_missing_reference(self): | CODE |
| LOW⚡ | …ge-desk/stages/02-group-and-triage/tests/test_stage.py | 35 | def test_action_evidence_uses_original_spans(self): | CODE |
| LOW⚡ | …ge-desk/stages/02-group-and-triage/tests/test_stage.py | 43 | def test_conflicts_and_unmatched_are_uncertain(self): | CODE |
| LOW⚡ | …ge-desk/stages/02-group-and-triage/tests/test_stage.py | 52 | def test_word_boundary_and_invalid_rule(self): | CODE |
| LOW⚡ | …ge-desk/stages/02-group-and-triage/tests/test_stage.py | 59 | def test_custom_phrase_can_classify_information(self): | CODE |
| LOW | …stages/04-streaming/tests/test_interrupted_receipts.py | 33 | def test_complete_malformed_response_preserves_prefix_and_explicit_failure(self): | CODE |
| LOW⚡ | …t-agent-shell/stages/04-streaming/tests/test_client.py | 34 | def test_quit_preserves_unanswered_ids(self): | CODE |
| LOW⚡ | …t-agent-shell/stages/04-streaming/tests/test_client.py | 41 | def test_budget_preserves_native_rejection_and_remaining_ids(self): | CODE |
| LOW⚡ | …t-agent-shell/stages/04-streaming/tests/test_client.py | 47 | def test_process_error_preserves_completed_prefix_and_all_ids(self): | CODE |
| LOW⚡ | …t-agent-shell/stages/04-streaming/tests/test_client.py | 53 | def test_empty_input_emits_no_events(self): | CODE |
| LOW⚡ | …t-agent-shell/stages/04-streaming/tests/test_client.py | 58 | def test_required_and_unknown_arguments_rejected(self): | CODE |
| LOW⚡ | …brary/stages/04-export-the-gallery/tests/test_stage.py | 37 | def test_html_is_portable_and_overlay_has_scaled_coordinates(self): | CODE |
| LOW⚡ | …brary/stages/04-export-the-gallery/tests/test_stage.py | 47 | def test_json_keeps_pixels_and_excludes_resolved_paths(self): | CODE |
| LOW⚡ | …brary/stages/04-export-the-gallery/tests/test_stage.py | 57 | def test_text_and_query_are_escaped(self): | CODE |
| LOW | …brary/stages/04-export-the-gallery/tests/test_stage.py | 68 | def test_unmatched_query_has_explicit_empty_state(self): | CODE |
| LOW | …brary/stages/04-export-the-gallery/tests/test_stage.py | 79 | def test_svg_script_element_is_rejected(self): | CODE |
| LOW | …brary/stages/04-export-the-gallery/tests/test_stage.py | 90 | def test_svg_event_and_external_reference_attributes_removed(self): | CODE |
| LOW | …ibrary/stages/01-validate-evidence/tests/test_stage.py | 36 | def test_valid_supplied_region_keeps_provenance(self): | CODE |
| LOW | …ibrary/stages/01-validate-evidence/tests/test_stage.py | 43 | def test_duplicate_asset_and_region_ids_rejected(self): | CODE |
| LOW | …ibrary/stages/01-validate-evidence/tests/test_stage.py | 55 | def test_negative_and_outside_boxes_rejected(self): | CODE |
| LOW | …ibrary/stages/01-validate-evidence/tests/test_stage.py | 70 | def test_escape_path_and_symlink_rejected(self): | CODE |
| LOW | …ibrary/stages/01-validate-evidence/tests/test_stage.py | 87 | def test_model_region_requires_explicit_review(self): | CODE |
| LOW | …ibrary/stages/01-validate-evidence/tests/test_stage.py | 97 | def test_missing_file_boolean_dimension_and_empty_text(self): | CODE |
| LOW | …/stages/03-review-vision-proposals/tests/test_stage.py | 17 | def test_text_and_labels_stay_separate(self): | CODE |
| LOW | …/stages/03-review-vision-proposals/tests/test_stage.py | 24 | def test_outside_and_nonfinite_coordinates_rejected(self): | CODE |
| LOW⚡ | …/stages/03-review-vision-proposals/tests/test_stage.py | 36 | def test_empty_text_and_nonstring_labels_rejected(self): | CODE |
| LOW⚡ | …/stages/03-review-vision-proposals/tests/test_stage.py | 46 | def test_empty_regions_are_not_invented(self): | CODE |
| LOW⚡ | …/stages/03-review-vision-proposals/tests/test_stage.py | 49 | def test_actual_request_serializes_image_and_model(self): | CODE |
| LOW | …/stages/03-review-vision-proposals/tests/test_stage.py | 77 | def test_bad_image_and_non_http_endpoint_rejected(self): | CODE |
| LOW⚡ | …rary/stages/02-search-text-regions/tests/test_stage.py | 33 | def test_unique_unicode_tokens(self): | CODE |
| 2704 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | scripts/translate_lessons.py | 165 | # ── NLLB-200: free, key-less, runs in the CI runner ────────────────────────── | COMMENT |
| MEDIUM | site/figures-capstone-g.js | 44 | // ── 24: plan-execute-replan — a failed step hands the cursor back ────────── | COMMENT |
| MEDIUM | site/figures-capstone-g.js | 135 | // ── 26: sandbox path jail — traversal bounces off the root fence ─────────── | COMMENT |
| MEDIUM | site/figures-capstone-g.js | 185 | // ── 29: end-to-end harness — one run token weaving through four strata ───── | COMMENT |
| MEDIUM | site/figures-capstone-g.js | 220 | // ── 41: eval pipeline — four gauges fold into one report ─────────────────── | COMMENT |
| MEDIUM | site/figures-capstone-g.js | 264 | // ── 48: DDP collectives — gradients meet in the middle, mean returns ─────── | COMMENT |
| MEDIUM | site/figures-capstone-g.js | 400 | // ── 52: experiment runner — two meters on the child, one kill path ───────── | COMMENT |
| MEDIUM | site/figures-capstone-g.js | 454 | // ── 53: result evaluator — per-seed differences carry the verdict ────────── | COMMENT |
| MEDIUM | site/figures-infra4.js | 58 | // ── i4-platform-lanes: PTU reserved lane vs shared on-demand lane ────────── | COMMENT |
| MEDIUM | site/figures-infra4.js | 97 | // ── i4-otel-glue: gateway spans fan out through OTel to two backends ─────── | COMMENT |
| MEDIUM | site/figures-infra4.js | 146 | // ── i4-canary-ramp: shadow mirror first, then gated traffic steps ────────── | COMMENT |
| MEDIUM | site/figures-infra4.js | 194 | // ── i4-incident-agents: supervisor fans out to agents, human gate acts ───── | COMMENT |
| MEDIUM | site/figures-nlp3.js | 32 | // ── pos-tagger: grammatical tags drop onto tokens one by one ──────────────── | COMMENT |
| MEDIUM | site/figures-nlp3.js | 56 | // ── dependency-arcs: labeled head→dependent arcs draw themselves ──────────── | COMMENT |
| MEDIUM | site/figures-nlp3.js | 80 | // ── qa-span: an answer span sweeps and highlights inside a passage ────────── | COMMENT |
| MEDIUM | site/figures-nlp3.js | 103 | // ── summarize-collapse: a long bar collapses into a short summary bar ──────── | COMMENT |
| MEDIUM | site/figures-nlp3.js | 129 | // ── topic-drift: scattered words drift into colored topic clusters ────────── | COMMENT |
| MEDIUM | site/figures-nlp3.js | 153 | // ── coref-links: pronoun mentions animate links back to one entity ────────── | COMMENT |
| MEDIUM | site/figures-nlp3.js | 176 | // ── nli-router: a premise/hypothesis pair routes to one of three labels ───── | COMMENT |
| MEDIUM | site/figures-nlp3.js | 233 | // ── constrained-decoder: invalid tokens get masked to -inf each step ──────── | COMMENT |
| MEDIUM | site/figures-systems3.js | 38 | // ── masked-diffusion-unmask (Show-o): masked grid fills in over steps ────── | COMMENT |
| MEDIUM | site/figures-systems3.js | 72 | // ── any-to-any-stream (MIO): four modality tokens merge into one stream ───── | COMMENT |
| MEDIUM | site/figures-systems3.js | 112 | // ── video-diffusion-denoise (Sora-style): noisy frame strip sharpens ─────── | COMMENT |
| MEDIUM | site/figures-systems3.js | 182 | // ── agentic-rag-loop: retrieve → reason → act cycle ──────────────────────── | COMMENT |
| MEDIUM | site/figures-systems3.js | 222 | // ── mcp-nxm-collapse: N hosts × M servers → one protocol hub ─────────────── | COMMENT |
| MEDIUM | site/figures-agents3.js | 29 | // ── htn-tree-expand: a task tree decomposing node by node ────────────────── | COMMENT |
| MEDIUM | site/figures-agents3.js | 68 | // ── workflow-chain: prompt chaining links lighting up in sequence ────────── | COMMENT |
| MEDIUM | site/figures-agents3.js | 98 | // ── actor-mailbox: messages flying into actor inboxes, async ─────────────── | COMMENT |
| MEDIUM | site/figures-agents3.js | 163 | // ── computer-use-cursor: a cursor gliding across a mock UI ───────────────── | COMMENT |
| MEDIUM | site/figures-agents3.js | 189 | // ── voice-pipeline: a waveform morphing into a text token ────────────────── | COMMENT |
| MEDIUM | site/figures-agents3.js | 226 | // ── injection-hijack: a malicious token glowing red, hijacking flow ──────── | COMMENT |
| MEDIUM | site/figures-agents3.js | 262 | // ── failure-cascade: an error tumbling down a chain of agents ────────────── | COMMENT |
| MEDIUM | site/figures-capstone-c.js | 79 | // ── 34: transformer block — two residual bypasses skip the sublayers ──────── | COMMENT |
| MEDIUM | site/figures-capstone-c.js | 173 | // ── 37: weight remapping — pretrained names rewired into local names ──────── | COMMENT |
| MEDIUM | site/figures-foundations2.js | 24 | // ── data-augmentation: one source image, four transformed copies ─────────── | COMMENT |
| MEDIUM | site/figures-foundations2.js | 87 | // ── transfer-learning: freeze a pretrained backbone, train the head ──────── | COMMENT |
| MEDIUM | site/figures-foundations2.js | 121 | // ── batchnorm-inference: batch stats in training vs running averages ─────── | COMMENT |
| MEDIUM | site/figures-foundations2.js | 160 | // ── ctc-collapse: per-frame chars collapse to a transcript ───────────────── | COMMENT |
| MEDIUM | site/figures-foundations2.js | 205 | // ── mfcc-pipeline: spectrogram → mel → log → DCT → keep N coeffs ──────────── | COMMENT |
| MEDIUM | site/figures-foundations2.js | 244 | // ── autoencoder-bottleneck: reconstruction quality vs latent dim ─────────── | COMMENT |
| MEDIUM | site/figures-foundations2.js | 297 | // ── normalizing-flow: invertible map, base → target, log-det Jacobian ────── | COMMENT |
| MEDIUM | site/figures-foundations2.js | 342 | // ── score-matching: score vector field and Langevin sampling steps ───────── | COMMENT |
| MEDIUM | site/figures-setup.js | 69 | // ── s0-env-stack: the four-layer dev stack assembles bottom-up ──────────── | COMMENT |
| MEDIUM | site/figures-setup.js | 95 | // ── s0-commit-dag: a commit DAG grows, a branch merges back to main ─────── | COMMENT |
| MEDIUM | site/figures-setup.js | 125 | // ── s0-gpu-dispatch: the same matmul batch on CPU vs GPU lanes ──────────── | COMMENT |
| MEDIUM | site/figures-setup.js | 168 | // ── s0-secret-inject: the key flows from .env into the request header ───── | COMMENT |
| MEDIUM | site/figures-setup.js | 422 | // ── s0-process-fork: the process tree forks down to your training run ───── | COMMENT |
| MEDIUM | site/figures-setup.js | 468 | // ── s0-flame-hot: the profile widens on the real bottleneck ─────────────── | COMMENT |
| MEDIUM | site/figures-frontier.js | 159 | // ── world-model-rollout: imagine future states with a learned model ──────── | COMMENT |
| MEDIUM | site/figures-frontier.js | 243 | // ── pass-at-k: pass@k = 1 - (1-p)^k rises toward 1 as k grows ─────────────── | COMMENT |
| MEDIUM | site/figures-frontier.js | 279 | // ── eval-harness-matrix: tasks x variants grid, aggregate per variant ────── | COMMENT |
| MEDIUM | site/figures-frontier.js | 332 | // ── canary-rollout: traffic split, error rate, rollback trigger ──────────── | COMMENT |
| MEDIUM | site/figures-alignment4.js | 87 | // ── bias: identical resumes, one scorer, unequal scores ───────────────────── | COMMENT |
| MEDIUM | site/figures-alignment4.js | 133 | // ── fairness: three criteria, unequal base rates, pick two ────────────────── | COMMENT |
| MEDIUM | site/figures-alignment4.js | 163 | // ── differential privacy: clip each gradient, then add calibrated noise ───── | COMMENT |
| MEDIUM | site/figures-alignment4.js | 212 | // ── watermarking: green-list bias at sampling, z-score at detection ───────── | COMMENT |
| MEDIUM | site/figures-alignment4.js | 250 | // ── EU AI Act: obligations arrive in dated waves ──────────────────────────── | COMMENT |
| MEDIUM | site/figures-alignment4.js | 281 | // ── EchoLeak: zero-click packet crosses the org trust boundary ────────────── | COMMENT |
| MEDIUM | site/figures-alignment4.js | 316 | // ── cards: dataset, model, and system scopes telescope outward ────────────── | COMMENT |
| MEDIUM | site/figures-alignment4.js | 381 | // ── moderation: input, model, output layers each get a turn ───────────────── | COMMENT |
| 657 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | .claude/skills/start-learning/SKILL.md | 105 | ## Step 1: The interview (3 questions, keep it short) | COMMENT |
| LOW | .claude/skills/start-learning/SKILL.md | 119 | ## Step 2: Placement | COMMENT |
| LOW | .claude/skills/start-learning/SKILL.md | 139 | ## Step 3: Write LEARNING.md | COMMENT |
| LOW | .claude/skills/start-learning/SKILL.md | 173 | ## Step 4: Hand off | COMMENT |
| LOW⚡ | .claude/skills/check-understanding/SKILL.md | 54 | ### Step 1: Resolve the Phase | COMMENT |
| LOW⚡ | .claude/skills/check-understanding/SKILL.md | 58 | ### Step 2: Read the Phase Content | COMMENT |
| LOW⚡ | .claude/skills/check-understanding/SKILL.md | 64 | ### Step 3: Generate 8 Questions | COMMENT |
| LOW | .claude/skills/check-understanding/SKILL.md | 86 | ### Step 4: Present Questions One at a Time | COMMENT |
| LOW | .claude/skills/check-understanding/SKILL.md | 110 | ### Step 5: Track and Score | COMMENT |
| LOW | .claude/skills/check-understanding/SKILL.md | 116 | ### Step 6: Show Results | COMMENT |
| LOW | .claude/skills/check-understanding/SKILL.md | 136 | ### Step 7: Wrong Answer Breakdown | COMMENT |
| LOW | .claude/skills/check-understanding/SKILL.md | 148 | ### Step 8: What Next? | COMMENT |
| LOW | .claude/skills/build-project/SKILL.md | 50 | ## Step 0: find or create progress | COMMENT |
| LOW | .claude/skills/build-project/SKILL.md | 77 | ## Step 1: set up the workspace (first stage only) | COMMENT |
| LOW | .claude/skills/build-project/SKILL.md | 90 | ## Step 2: teach the stage | COMMENT |
| LOW | .claude/skills/build-project/SKILL.md | 119 | ## Step 3: record and point forward | COMMENT |
| LOW | manuals/a2a-101/research/sources/specification.md | 1847 | ### Step 2: Client obtains credentials (out-of-band OAuth 2.0 flow) | COMMENT |
| LOW | manuals/a2a-101/research/sources/specification.md | 1849 | ### Step 3: Client fetches authenticated extended Agent Card | COMMENT |
| LOW | manuals/a2a-101/research/sources/specification-main.md | 1847 | ### Step 2: Client obtains credentials (out-of-band OAuth 2.0 flow) | COMMENT |
| LOW | manuals/a2a-101/research/sources/specification-main.md | 1849 | ### Step 3: Client fetches authenticated extended Agent Card | COMMENT |
| LOW | …01/sections/1-2-docker-agent-run-sandbox-end-to-end.md | 42 | ## Step 1: allow the model host | COMMENT |
| LOW | …01/sections/1-2-docker-agent-run-sandbox-end-to-end.md | 60 | ## Step 2: keep the state directories in the workspace | COMMENT |
| LOW | …01/sections/1-2-docker-agent-run-sandbox-end-to-end.md | 75 | ## Step 3: the launch summary | COMMENT |
| LOW | …01/sections/1-2-docker-agent-run-sandbox-end-to-end.md | 106 | ## Step 4: an ordinary sandbox | COMMENT |
| LOW | …01/sections/1-2-docker-agent-run-sandbox-end-to-end.md | 110 | ## Step 5: the session store fails | COMMENT |
| LOW | …01/sections/1-2-docker-agent-run-sandbox-end-to-end.md | 128 | ## Step 6: the model call from inside | COMMENT |
| LOW | …01/sections/1-2-docker-agent-run-sandbox-end-to-end.md | 167 | ## Step 7: remove the sandbox | COMMENT |
| LOW⚡ | …lessons/24-rag-retrieval-and-data-pipelines/docs/en.md | 234 | ### Step 1: Normalize Tokens | COMMENT |
| LOW⚡ | …lessons/24-rag-retrieval-and-data-pipelines/docs/en.md | 240 | ### Step 2: Chunk With Stable Identity | COMMENT |
| LOW⚡ | …lessons/24-rag-retrieval-and-data-pipelines/docs/en.md | 246 | ### Step 3: Exclude Inactive Sources Before Indexing | COMMENT |
| LOW⚡ | …lessons/24-rag-retrieval-and-data-pipelines/docs/en.md | 252 | ### Step 4: Score Transparently | COMMENT |
| LOW⚡ | …lessons/24-rag-retrieval-and-data-pipelines/docs/en.md | 258 | ### Step 5: Return Provenance | COMMENT |
| LOW⚡ | …lessons/24-rag-retrieval-and-data-pipelines/docs/en.md | 264 | ### Step 6: Evaluate the Retriever | COMMENT |
| LOW | …ons/04-context-knowledge-memory-and-caching/docs/en.md | 148 | ### Step 1: Inventory the context | COMMENT |
| LOW | …ons/04-context-knowledge-memory-and-caching/docs/en.md | 164 | ### Step 2: Create a source registry | COMMENT |
| LOW | …ons/04-context-knowledge-memory-and-caching/docs/en.md | 182 | ### Step 3: Design retrieval with abstention | COMMENT |
| LOW | …ons/04-context-knowledge-memory-and-caching/docs/en.md | 195 | ### Step 4: Budget the prompt | COMMENT |
| LOW | …ons/04-context-knowledge-memory-and-caching/docs/en.md | 207 | ### Step 5: Establish maintenance | COMMENT |
| LOW⚡ | …s/32-architect-professional-system-capstone/docs/en.md | 225 | ### Step 1: Encode Requirements | COMMENT |
| LOW⚡ | …s/32-architect-professional-system-capstone/docs/en.md | 230 | ### Step 2: Encode Decisions | COMMENT |
| LOW⚡ | …s/32-architect-professional-system-capstone/docs/en.md | 236 | ### Step 3: Encode Controls | COMMENT |
| LOW⚡ | …s/32-architect-professional-system-capstone/docs/en.md | 242 | ### Step 4: Evaluate Gates | COMMENT |
| LOW⚡ | …s/32-architect-professional-system-capstone/docs/en.md | 248 | ### Step 5: Make the Release Decision | COMMENT |
| LOW⚡ | …6-production-observability-latency-and-cost/docs/en.md | 243 | ### Step 1: Represent One Task Trajectory | COMMENT |
| LOW⚡ | …6-production-observability-latency-and-cost/docs/en.md | 250 | ### Step 2: Aggregate Without Hiding Failure | COMMENT |
| LOW⚡ | …6-production-observability-latency-and-cost/docs/en.md | 256 | ### Step 3: Compare Variants | COMMENT |
| LOW⚡ | …6-production-observability-latency-and-cost/docs/en.md | 261 | ### Step 4: Evaluate Service Objectives | COMMENT |
| LOW | …lessons/05-output-evaluation-and-validation/docs/en.md | 191 | ### Step 1: Define release gates | COMMENT |
| LOW | …lessons/05-output-evaluation-and-validation/docs/en.md | 203 | ### Step 2: Build a validation record | COMMENT |
| LOW | …lessons/05-output-evaluation-and-validation/docs/en.md | 239 | ### Step 3: Separate generation and review | COMMENT |
| LOW | …lessons/05-output-evaluation-and-validation/docs/en.md | 253 | ### Step 4: Calibrate graders | COMMENT |
| LOW | …lessons/05-output-evaluation-and-validation/docs/en.md | 259 | ### Step 5: Close the loop | COMMENT |
| LOW | …/31-architect-foundations-scenario-capstone/docs/en.md | 153 | ### Step 1: Choose one primary lens | COMMENT |
| LOW | …/31-architect-foundations-scenario-capstone/docs/en.md | 170 | ### Step 2: Complete the architecture packet | COMMENT |
| LOW | …/31-architect-foundations-scenario-capstone/docs/en.md | 183 | ### Step 3: Encode the packet as JSON | COMMENT |
| LOW | …/31-architect-foundations-scenario-capstone/docs/en.md | 212 | ### Step 4: Run failure-first tests | COMMENT |
| LOW | …/31-architect-foundations-scenario-capstone/docs/en.md | 232 | ### Step 5: Transfer across all six lenses | COMMENT |
| LOW | …/31-architect-foundations-scenario-capstone/docs/en.md | 258 | ### Step 6: Defend tradeoffs | COMMENT |
| LOW | …/31-architect-foundations-scenario-capstone/docs/en.md | 268 | ### Step 7: Conduct independent review | COMMENT |
| LOW | …lessons/03-prompting-and-task-decomposition/docs/en.md | 147 | ### Step 1: Write the acceptance card | COMMENT |
| 1202 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …planner/stages/01-estimate-cost/tests/test_stage_01.py | 1 | CODE | |
| LOW | …planner/stages/01-estimate-cost/tests/test_stage_01.py | 1 | CODE | |
| LOW | …planner/stages/01-estimate-cost/tests/test_stage_01.py | 1 | CODE | |
| LOW | …planner/stages/01-estimate-cost/tests/test_stage_01.py | 2 | CODE | |
| LOW | …planner/stages/01-estimate-cost/tests/test_stage_01.py | 3 | CODE | |
| LOW | …/04-schedule-under-deadlines/tests/test_integration.py | 1 | CODE | |
| LOW | …/04-schedule-under-deadlines/tests/test_integration.py | 1 | CODE | |
| LOW | …/04-schedule-under-deadlines/tests/test_integration.py | 1 | CODE | |
| LOW | …/04-schedule-under-deadlines/tests/test_integration.py | 1 | CODE | |
| LOW | …ges/04-schedule-under-deadlines/tests/test_stage_04.py | 1 | CODE | |
| LOW | …ges/04-schedule-under-deadlines/tests/test_stage_04.py | 1 | CODE | |
| LOW | …ges/04-schedule-under-deadlines/tests/test_stage_04.py | 1 | CODE | |
| LOW | …ges/04-schedule-under-deadlines/tests/test_stage_04.py | 2 | CODE | |
| LOW | …ges/04-schedule-under-deadlines/tests/test_stage_04.py | 3 | CODE | |
| LOW | …nner/stages/02-reserve-capacity/tests/test_stage_02.py | 1 | CODE | |
| LOW | …nner/stages/02-reserve-capacity/tests/test_stage_02.py | 1 | CODE | |
| LOW | …nner/stages/02-reserve-capacity/tests/test_stage_02.py | 1 | CODE | |
| LOW | …nner/stages/02-reserve-capacity/tests/test_stage_02.py | 2 | CODE | |
| LOW | …nner/stages/02-reserve-capacity/tests/test_stage_02.py | 3 | CODE | |
| LOW | …er/stages/03-settle-and-release/tests/test_stage_03.py | 1 | CODE | |
| LOW | …er/stages/03-settle-and-release/tests/test_stage_03.py | 1 | CODE | |
| LOW | …er/stages/03-settle-and-release/tests/test_stage_03.py | 1 | CODE | |
| LOW | …er/stages/03-settle-and-release/tests/test_stage_03.py | 2 | CODE | |
| LOW | …er/stages/03-settle-and-release/tests/test_stage_03.py | 3 | CODE | |
| LOW | projects/agent-budget-planner/solution/demo.py | 2 | CODE | |
| LOW | projects/inbox-triage-desk/solution/main.py | 1 | CODE | |
| LOW | projects/rust-agent-shell/solution/demo.py | 2 | CODE | |
| LOW | projects/rust-agent-shell/solution/demo.py | 2 | CODE | |
| LOW | projects/rust-agent-shell/solution/demo.py | 2 | CODE | |
| LOW | …brary/stages/04-export-the-gallery/tests/test_stage.py | 1 | CODE | |
| LOW | …ibrary/stages/01-validate-evidence/tests/test_stage.py | 2 | CODE | |
| LOW | projects/visual-evidence-library/solution/main.py | 1 | CODE | |
| LOW | …proving-skill-loop/stages/02-skill/tests/test_skill.py | 2 | CODE | |
| LOW | …proving-skill-loop/stages/02-skill/tests/test_skill.py | 3 | CODE | |
| LOW | …proving-skill-loop/stages/02-skill/tests/test_skill.py | 4 | CODE | |
| LOW | …-improving-skill-loop/stages/02-skill/starter/skill.py | 8 | CODE | |
| LOW | …ing-skill-loop/stages/03-propose/tests/test_propose.py | 2 | CODE | |
| LOW | …ing-skill-loop/stages/03-propose/tests/test_propose.py | 3 | CODE | |
| LOW | …ing-skill-loop/stages/03-propose/tests/test_propose.py | 4 | CODE | |
| LOW | …roving-skill-loop/stages/03-propose/starter/propose.py | 8 | CODE | |
| LOW | …roving-skill-loop/stages/03-propose/starter/propose.py | 10 | CODE | |
| LOW | …roving-skill-loop/stages/03-propose/starter/propose.py | 12 | CODE | |
| LOW | …ing-skill-loop/stages/01-dataset/tests/test_dataset.py | 2 | CODE | |
| LOW | …ing-skill-loop/stages/01-dataset/tests/test_dataset.py | 3 | CODE | |
| LOW | …ing-skill-loop/stages/01-dataset/tests/test_dataset.py | 4 | CODE | |
| LOW | …roving-skill-loop/stages/01-dataset/starter/dataset.py | 8 | CODE | |
| LOW | …skill-loop/stages/04-promotion/tests/test_promotion.py | 2 | CODE | |
| LOW | …skill-loop/stages/04-promotion/tests/test_promotion.py | 3 | CODE | |
| LOW | …skill-loop/stages/04-promotion/tests/test_promotion.py | 4 | CODE | |
| LOW | …ng-skill-loop/stages/04-promotion/starter/promotion.py | 8 | CODE | |
| LOW | …ng-skill-loop/stages/04-promotion/starter/promotion.py | 10 | CODE | |
| LOW | …ng-skill-loop/stages/04-promotion/starter/promotion.py | 12 | CODE | |
| LOW | projects/self-improving-skill-loop/solution/demo.py | 2 | CODE | |
| LOW | projects/self-improving-skill-loop/solution/demo.py | 2 | CODE | |
| LOW | projects/self-improving-skill-loop/solution/demo.py | 2 | CODE | |
| LOW | projects/self-improving-skill-loop/solution/demo.py | 2 | CODE | |
| LOW | projects/self-improving-skill-loop/solution/demo.py | 6 | CODE | |
| LOW | …oryboard-builder/stages/01-roles/tests/test_stage_1.py | 1 | CODE | |
| LOW | …oryboard-builder/stages/01-roles/tests/test_stage_1.py | 1 | CODE | |
| LOW | …oryboard-builder/stages/01-roles/tests/test_stage_1.py | 1 | CODE | |
| 1443 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | projects/rag-freshness-pipeline/README.md | 1 | # RAG Freshness Pipeline | COMMENT |
| MEDIUM⚡ | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 4818 | - [#2322](https://github.com/docker/docker-agent/pull/2322) - fix: resolve duplicate RAG tool names and nil pointer pani | CODE |
| MEDIUM⚡ | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 4820 | - [#2324](https://github.com/docker/docker-agent/pull/2324) - fix: pass Models map to RAG ManagersBuildConfig for model | CODE |
| MEDIUM⚡ | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 4828 | - [#2344](https://github.com/docker/docker-agent/pull/2344) - feat: support custom providers in RAG embedding and rerank | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 322 | - [#4418](https://github.com/docker/docker-agent/pull/4418) - fix(acp): separate root context from multi-agent cost | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 961 | - [#4097](https://github.com/docker/docker-agent/pull/4097) - fix(#4097): propagate sustained RAG 5xx/408 indexing failu | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 976 | - [#4151](https://github.com/docker/docker-agent/pull/4151) - fix(#4097): propagate sustained RAG 5xx/408 indexing failu | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 2313 | - [#3451](https://github.com/docker/docker-agent/pull/3451) - feat: add docker agent board, a Kanban TUI for orchestrati | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 2564 | - [#3393](https://github.com/docker/docker-agent/pull/3393) - fix: clamp max_tokens to the context window for OpenAI-com | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 2646 | - [#3326](https://github.com/docker/docker-agent/pull/3326) - refactor: replace test-mutated globals with dependency inj | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 2854 | - [#3227](https://github.com/docker/docker-agent/pull/3227) - feat: add plan builtin toolset for shared multi-agent coll | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 2946 | - [#3227](https://github.com/docker/docker-agent/pull/3227) - feat: add plan builtin toolset for shared multi-agent coll | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 3690 | - [#2876](https://github.com/docker/docker-agent/pull/2876) - docs: document mcp_catalog toolset and reorganize RAG refe | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 4028 | - [#2723](https://github.com/docker/docker-agent/pull/2723) - remote-runtime: close silent gaps, consolidate Runtime, sc | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 4028 | - [#2723](https://github.com/docker/docker-agent/pull/2723) - remote-runtime: close silent gaps, consolidate Runtime, sc | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 4227 | - [#2672](https://github.com/docker/docker-agent/pull/2672) - add js/wasm browser build with OpenRouter PKCE, agentic lo | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 4227 | - [#2672](https://github.com/docker/docker-agent/pull/2672) - add js/wasm browser build with OpenRouter PKCE, agentic lo | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 4351 | - [#2523](https://github.com/docker/docker-agent/pull/2523) - refactor(runtime): consolidate hook orchestration and cach | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 4363 | - [#2535](https://github.com/docker/docker-agent/pull/2535) - refactor(runtime): extract sub-session orchestration | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 5021 | - [#2210](https://github.com/docker/docker-agent/pull/2210) - Refactor RAG from agent-level config to standard toolset t | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 5067 | - [#2205](https://github.com/docker/docker-agent/pull/2205) - Simplify the runtime related RAG code a bit | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 5620 | - [#1973](https://github.com/docker/docker-agent/pull/1973) - Fix context window overflow: auto-recovery and proactive c | CODE |
| MEDIUM | …ndboxes-101/research/sources/docker-agent-CHANGELOG.md | 5708 | - [#1899](https://github.com/docker/docker-agent/pull/1899) - fix: Esc key not interrupting sub-agents in multi-agent se | CODE |
| MEDIUM | …er-sandboxes-101/research/sources/docs-model-runner.md | 1618 | [Claude Code](https://claude.com/product/claude-code) is [Anthropic's](https://www.anthropic.com/) command-line tool for | CODE |
| MEDIUM | …/6-5-dmr-compose-models-and-the-mcp-gateway-service.md | 52 | The 8B tag `ai/qwen3:latest` failed to pull on the recording Mac, as [docker-agent run, new, and doctor](#s-docker-agent | CODE |
| MEDIUM | site/figures-systems3.js | 182 | // ── agentic-rag-loop: retrieve → reason → act cycle ──────────────────────── | COMMENT |
| MEDIUM | site/figures-autoswarm5.js | 61 | // ── a5-scaffold-delta: one model feeds two scaffolds; the score bars land | COMMENT |
| MEDIUM | site/learning-paths.html | 144 | <a href="#career-route-agentic-ai-engineer" data-career-choice="agentic-ai-engineer"><span>Engineer tool-using | CODE |
| MEDIUM | site/learning-paths.html | 144 | <a href="#career-route-agentic-ai-engineer" data-career-choice="agentic-ai-engineer"><span>Engineer tool-using | CODE |
| MEDIUM | site/learning-paths.html | 231 | <div class="career-guide-footer"><p>Complete the LLM and tool-interface baseline first. Then treat every age | CODE |
| MEDIUM⚡ | site/glossary.html | 26 | <script type="application/ld+json">{"@context":"https://schema.org","@type":"DefinedTermSet","@id":"https://aiengineer | CODE |
| MEDIUM⚡ | site/glossary.html | 26 | <script type="application/ld+json">{"@context":"https://schema.org","@type":"DefinedTermSet","@id":"https://aiengineer | CODE |
| MEDIUM⚡ | site/glossary.html | 26 | <script type="application/ld+json">{"@context":"https://schema.org","@type":"DefinedTermSet","@id":"https://aiengineer | CODE |
| MEDIUM⚡ | site/glossary.html | 26 | <script type="application/ld+json">{"@context":"https://schema.org","@type":"DefinedTermSet","@id":"https://aiengineer | CODE |
| MEDIUM⚡ | site/glossary.html | 26 | <script type="application/ld+json">{"@context":"https://schema.org","@type":"DefinedTermSet","@id":"https://aiengineer | CODE |
| MEDIUM⚡ | site/glossary.html | 26 | <script type="application/ld+json">{"@context":"https://schema.org","@type":"DefinedTermSet","@id":"https://aiengineer | CODE |
| MEDIUM⚡ | site/glossary.html | 26 | <script type="application/ld+json">{"@context":"https://schema.org","@type":"DefinedTermSet","@id":"https://aiengineer | CODE |
| MEDIUM | site/glossary.html | 988 | <section aria-label="Prompting & context"><h3>Prompting & context</h3><ul><li><a href="glossary/chai | CODE |
| MEDIUM | glossary/myths.md | 69 | ## "Bigger context window = better" | COMMENT |
| MEDIUM | glossary/terms.md | 1544 | ### RAG (Retrieval-Augmented Generation) | COMMENT |
| MEDIUM | …rtifications/claude/assessments/ccdv-f/diagnostic.json | 317 | "https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks", | CODE |
| MEDIUM | certifications/claude/assessments/ccdv-f/mock-01.json | 488 | "https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks", | CODE |
| MEDIUM | certifications/claude/assessments/ccdv-f/mock-01.json | 865 | "https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks", | CODE |
| MEDIUM | certifications/claude/assessments/ccdv-f/mock-01.json | 931 | "https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks", | CODE |
| MEDIUM | certifications/claude/assessments/ccdv-f/mock-01.json | 975 | "https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks", | CODE |
| MEDIUM | certifications/claude/assessments/ccdv-f/mock-01.json | 997 | "https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks", | CODE |
| MEDIUM | certifications/claude/assessments/ccdv-f/mock-01.json | 1018 | "https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks", | CODE |
| MEDIUM | …rtifications/claude/assessments/ccao-f/diagnostic.json | 70 | "https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/reduce-hallucinations" | CODE |
| MEDIUM | certifications/claude/assessments/ccao-f/mock-01.json | 177 | "https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/reduce-hallucinations" | CODE |
| MEDIUM | certifications/claude/assessments/ccao-f/mock-01.json | 241 | "https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/reduce-hallucinations" | CODE |
| MEDIUM | certifications/claude/assessments/ccao-f/mock-01.json | 1051 | "https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/reduce-prompt-leak" | CODE |
| MEDIUM | certifications/claude/assessments/ccar-p/mock-01.json | 961 | "https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks", | CODE |
| MEDIUM | certifications/claude/assessments/ccar-p/mock-01.json | 1124 | "https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks", | CODE |
| MEDIUM | …lessons/24-rag-retrieval-and-data-pipelines/docs/en.md | 1 | # RAG, Retrieval, and Data Pipelines | COMMENT |
| MEDIUM | …lessons/24-rag-retrieval-and-data-pipelines/docs/en.md | 34 | ### RAG Is a Data System | COMMENT |
| MEDIUM | …4-evals-testing-debugging-and-observability/docs/en.md | 331 | - [Create strong empirical evaluations](https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/incr | CODE |
| MEDIUM | …s/32-architect-professional-system-capstone/docs/en.md | 96 | ### 5. Knowledge and RAG Design | COMMENT |
| MEDIUM | …ystem-capstone/outputs/architecture-packet-template.md | 37 | ## 5. RAG and Knowledge | COMMENT |
| MEDIUM | …stem-capstone/outputs/reference-architecture-packet.md | 34 | ## 5. RAG and Knowledge | COMMENT |
| MEDIUM | …lessons/05-output-evaluation-and-validation/docs/en.md | 351 | - [Anthropic: Reduce hallucinations](https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/reduce- | CODE |
| 98 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | …-notes-to-actions/stages/01-parse-lines/starter/cli.py | 98 | + '</pre><script>document.getElementById("download").onclick=()=>{const decisions=Object.fromEntries([...documen | CODE |
| HIGH | projects/meeting-notes-to-actions/solution/cli.py | 98 | + '</pre><script>document.getElementById("download").onclick=()=>{const decisions=Object.fromEntries([...documen | CODE |
| HIGH | …kshop-materials-planner/stages/01-model/starter/cli.py | 45 | + ';const slider=document.querySelector("#participants");function update(){const rows=scenarios[slider.value];docume | CODE |
| HIGH | projects/workshop-materials-planner/solution/cli.py | 45 | + ';const slider=document.querySelector("#participants");function update(){const rows=scenarios[slider.value];docume | CODE |
| HIGH | projects/document-extraction-desk/solution/main.py | 133 | script = """const data=JSON.parse(document.querySelector('#source-data').textContent);const view=document.querySelec | CODE |
| HIGH | …/stages/01-parse-the-capability-request/starter/cli.py | 33 | 'if touch /probe 2>/dev/null; then echo root_write=allowed; exit 2; else echo root_write=denied; fi; cat /proc/n | CODE |
| HIGH | projects/sandbox-ladder/solution/cli.py | 33 | 'if touch /probe 2>/dev/null; then echo root_write=allowed; exit 2; else echo root_write=denied; fi; cat /proc/n | CODE |
| HIGH | manuals/docker-sandboxes-101/capture/run.py | 1046 | commit = f"printf hi > from-sandbox.txt && git add from-sandbox.txt && printf 'from sandbox\\n' > /tmp/msg && {INSID | CODE |
| HIGH | manuals/docker-sandboxes-101/capture/run.py | 1686 | probe = "cd \"$PWD\" && mkdir -p .m101/probe && python3 -c 'import sqlite3; c = sqlite3.connect(\".m101/probe/x. | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 529 | outputs, _ = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 538 | outputs, _ = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 547 | outputs, _ = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 553 | outputs, _ = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 559 | outputs, _ = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 566 | outputs, _ = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 577 | outputs, _ = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 586 | outputs, _ = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 595 | outputs, log = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 604 | outputs, _ = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 614 | outputs, _ = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 622 | outputs, _ = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 630 | outputs, _ = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 646 | outputs, log = self.push(before=before, forced=forced) | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 657 | outputs, log = self.push() | CODE |
| HIGH⚡ | scripts/test_translate_workflow.py | 670 | outputs, _ = self.push(cwd=shallow) | CODE |
| HIGH | …cations/mcpa/lessons/03-json-rpc-and-meta/code/main.py | 203 | "violation": "a request id must never be null, even when every other field is well formed", | CODE |
| HIGH⚡ | …egistry-gateways-and-sdk-tiers/code/tests/test_main.py | 34 | for bad in ("^1.2.3", "~1.2.3", ">=1.2.3", "<=1.2.3", "1.x", "1.2.*", "1.2 || 1.3", "1 - 2"): | CODE |
| HIGH | …ering/01-prompt-engineering/code/prompt_engineering.py | 418 | 'Output: {"sentiment": "negative", "food": null, "service": "negative"}' | CODE |
| HIGH | phases/13-tools-and-protocols/14-mcp-apps/code/main.py | 97 | if (event.origin !== hostOrigin || !event.data || event.data.jsonrpc !== "2.0") return; | CODE |
| HIGH | phases/13-tools-and-protocols/14-mcp-apps/code/main.py | 98 | if (event.data.id === 0 && event.data.result) {{ | CODE |
| HIGH | …00-setup-and-tooling/01-dev-environment/code/verify.py | 111 | return "Run `sudo apt-get update && sudo apt-get install -y git`, then `git --version`." | CODE |
| HIGH | …00-setup-and-tooling/01-dev-environment/code/verify.py | 124 | "Run `fnm install 22 && fnm use 22`, then `node --version`.", | CODE |
| HIGH | …formers-deep-dive/16-speculative-decoding/code/main.py | 136 | print("=== acceptance rate vs KL(q || p) ===") | CODE |
| HIGH⚡ | …h-foundations/14-norms-and-distances/code/distances.py | 375 | print(f" KL(P || Q) = {kl_pq:.4f} nats") | CODE |
| HIGH⚡ | …h-foundations/14-norms-and-distances/code/distances.py | 376 | print(f" KL(Q || P) = {kl_qp:.4f} nats") | CODE |
| HIGH⚡ | …h-foundations/14-norms-and-distances/code/distances.py | 386 | print(f" KL(P || Q) = {kl_divergence(p2, q2):.4f} nats") | CODE |
| HIGH⚡ | …h-foundations/14-norms-and-distances/code/distances.py | 387 | print(f" KL(Q || P) = {kl_divergence(q2, p2):.4f} nats") | CODE |
| HIGH | …tions/09-information-theory/code/information_theory.py | 225 | print(f" KL(true || good): {kl_good:.4f} bits") | CODE |
| HIGH | …tions/09-information-theory/code/information_theory.py | 226 | print(f" KL(true || bad): {kl_bad:.4f} bits") | CODE |
| HIGH | …tions/09-information-theory/code/information_theory.py | 240 | print(f" KL(P || Q) = {kl_divergence(p, q):.4f} bits") | CODE |
| HIGH | …tions/09-information-theory/code/information_theory.py | 241 | print(f" KL(Q || P) = {kl_divergence(q, p):.4f} bits") | CODE |
| HIGH | …one-projects/05-autonomous-research-agent/code/main.py | 136 | tree.push(root) | CODE |
| HIGH⚡ | …one-projects/05-autonomous-research-agent/code/main.py | 160 | tree.push(ch) | CODE |
| HIGH | …s-systems/10-claude-code-permission-modes/code/main.py | 161 | Action("shell", "git add /tmp/secrets.txt && git push"), | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …desk/stages/03-draft-for-review/tests/test_endpoint.py | 76 | CODE | |
| LOW | projects/visual-evidence-library/solution/main.py | 211 | CODE | |
| LOW | projects/self-improving-skill-loop/solution/propose.py | 13 | CODE | |
| LOW | …rness/stages/01-validate-predictions/starter/record.py | 16 | CODE | |
| LOW | projects/local-model-eval-harness/solution/record.py | 16 | CODE | |
| LOW | projects/subtitle-localization-desk/solution/main.py | 61 | CODE | |
| LOW | projects/csv-repair-workbench/solution/main.py | 54 | CODE | |
| LOW | …r-pipeline/stages/03-transcribe/tests/test_endpoint.py | 72 | CODE | |
| LOW | …voice-note-transcriber-pipeline/solution/transcribe.py | 12 | CODE | |
| LOW | projects/pr-review-reporter/solution/diff_parser.py | 35 | CODE | |
| LOW | projects/browser-agent/solution/pixels.py | 8 | CODE | |
| LOW | projects/sensitive-text-redaction-gate/solution/main.py | 13 | CODE | |
| LOW | projects/prompt-regression-tester/solution/main.py | 34 | CODE | |
| LOW | projects/prompt-regression-tester/solution/main.py | 61 | CODE | |
| LOW | projects/csv-sql-question-workbench/solution/main.py | 12 | CODE | |
| LOW | projects/csv-sql-question-workbench/solution/main.py | 80 | CODE | |
| LOW | projects/mcp-at-scale/solution/protocol.py | 15 | CODE | |
| LOW | …shop-materials-planner/stages/01-model/starter/main.py | 15 | CODE | |
| LOW | projects/workshop-materials-planner/solution/main.py | 15 | CODE | |
| LOW | …nt/stages/02-apply-exact-patches/tests/test_cleanup.py | 11 | CODE | |
| LOW | projects/tiny-coding-agent/solution/main.py | 105 | CODE | |
| LOW | projects/infrastructure-plan-explainer/solution/main.py | 125 | CODE | |
| LOW | projects/document-extraction-desk/solution/main.py | 65 | CODE | |
| LOW | manuals/a2a-101/capture/run.py | 276 | CODE | |
| LOW | manuals/a2a-101/capture/wire.py | 39 | CODE | |
| LOW | manuals/a2a-101/capture/a2a_ref.py | 264 | CODE | |
| LOW | manuals/a2a-101/capture/a2a_ref.py | 320 | CODE | |
| LOW | manuals/a2a-101/capture/a2a_ref.py | 510 | CODE | |
| LOW | scripts/build_readme_i18n.py | 45 | CODE | |
| LOW | scripts/build_readme_i18n.py | 158 | CODE | |
| LOW | scripts/link_check.py | 115 | CODE | |
| LOW | scripts/link_check.py | 278 | CODE | |
| LOW | scripts/debias_quizzes.py | 118 | CODE | |
| LOW | scripts/audit_readme_locales.py | 355 | CODE | |
| LOW | scripts/check_mcpa_wire.py | 115 | CODE | |
| LOW | scripts/check_mcpa_wire.py | 143 | CODE | |
| LOW | scripts/check_mcpa_wire.py | 188 | CODE | |
| LOW | scripts/build_book.py | 109 | CODE | |
| LOW | scripts/check_readme_counts.py | 198 | CODE | |
| LOW | scripts/install_skills.py | 90 | CODE | |
| LOW | scripts/install_skills.py | 390 | CODE | |
| LOW | scripts/test_skill_artifact_bundles.py | 160 | CODE | |
| LOW | scripts/test_skill_artifact_bundles.py | 344 | CODE | |
| LOW | scripts/test_skill_artifact_bundles.py | 394 | CODE | |
| LOW | scripts/audit_certifications.py | 397 | CODE | |
| LOW | scripts/audit_certifications.py | 617 | CODE | |
| LOW | scripts/audit_certifications.py | 741 | CODE | |
| LOW | scripts/backfill_certification_references.py | 62 | CODE | |
| LOW | scripts/project_test.py | 145 | CODE | |
| LOW | scripts/project_test.py | 294 | CODE | |
| LOW | scripts/translate_ui_strings.py | 63 | CODE | |
| LOW | scripts/translate_ui_strings.py | 164 | CODE | |
| LOW | …s/04-context-knowledge-memory-and-caching/code/main.py | 18 | CODE | |
| LOW | …ssons/05-output-evaluation-and-validation/code/main.py | 24 | CODE | |
| LOW | …1-architect-foundations-scenario-capstone/code/main.py | 89 | CODE | |
| LOW | …sons/11-mcp-server-design-and-integration/code/main.py | 569 | CODE | |
| LOW | …ude/lessons/12-claude-agent-sdk-and-hooks/code/main.py | 46 | CODE | |
| LOW | …ude/lessons/12-claude-agent-sdk-and-hooks/code/main.py | 129 | CODE | |
| LOW | …08-messages-api-and-application-lifecycle/code/main.py | 227 | CODE | |
| LOW | …9-structured-output-and-defensive-parsing/code/main.py | 51 | CODE | |
| 191 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | …budget-planner/stages/01-estimate-cost/starter/main.py | 0 | fill in the stage interfaces in your own workspace. | STRING |
| HIGH | …harness/stages/01-validate-predictions/starter/main.py | 0 | fill in the stage interfaces in your own workspace. | STRING |
| HIGH | …notes-to-actions/stages/01-parse-lines/starter/main.py | 0 | fill in the stage interfaces in your own workspace. | STRING |
| HIGH | …notes-search/stages/01-normalize-notes/starter/main.py | 0 | fill in the stage interfaces in your own workspace. | STRING |
| HIGH | …ession-tester/stages/01-validate-cases/starter/main.py | 0 | fill in the stage interfaces in your own workspace. | STRING |
| HIGH | …uation-lab/stages/01-validate-rankings/starter/main.py | 0 | fill in the stage interfaces in your own workspace. | STRING |
| HIGH | …-auditor/stages/01-fingerprint-records/starter/main.py | 0 | fill in the stage interfaces in your own workspace. | STRING |
| HIGH | …y-coding-agent/stages/01-confine-paths/starter/main.py | 0 | fill in the stage interfaces in your own workspace. | STRING |
| HIGH | …yboard-builder/stages/01-roles/starter/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …ects/chart-storyboard-builder/solution/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …calization-desk/stages/01-cues/starter/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …ts/subtitle-localization-desk/solution/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …-sort-desk/stages/01-inventory/starter/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …cts/download-folder-sort-desk/solution/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …ir-workbench/stages/01-profile/starter/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | projects/csv-repair-workbench/solution/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …-path-finder/stages/01-catalog/starter/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …ommunity-learning-path-finder/solution/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …record-linker/stages/01-import/starter/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | projects/catalog-record-linker/solution/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …n-rehearsal/stages/01-snapshot/starter/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …/database-migration-rehearsal/solution/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …erials-planner/stages/01-model/starter/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …ts/workshop-materials-planner/solution/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …-plan-explainer/stages/01-load/starter/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | …infrastructure-plan-explainer/solution/presentation.py | 0 | write standalone accessible html. review data is downloadable and edited as json. | STRING |
| HIGH | phases/11-llm-engineering/07-advanced-rag/code/main.py | 0 | acme corp refund policy. all standard plan customers are eligible for a full refund within 30 days of purchase. enterpri | STRING |
| HIGH | phases/11-llm-engineering/06-rag/code/main.py | 0 | acme corp refund policy. all standard plan customers are eligible for a full refund within 30 days of purchase. enterpri | STRING |
| HIGH | …es/11-llm-engineering/04-embeddings/code/embeddings.py | 0 | acme corp refund policy. all standard plan customers are eligible for a full refund within 30 days of purchase. enterpri | STRING |
| HIGH | phases/11-llm-engineering/07-advanced-rag/code/main.py | 0 | acme corp product overview. acme corp offers three product tiers: starter, professional, and enterprise. the starter pla | STRING |
| HIGH | phases/11-llm-engineering/06-rag/code/main.py | 0 | acme corp product overview. acme corp offers three product tiers: starter, professional, and enterprise. the starter pla | STRING |
| HIGH | …es/11-llm-engineering/04-embeddings/code/embeddings.py | 0 | acme corp product overview. acme corp offers three product tiers: starter, professional, and enterprise. the starter pla | STRING |
| HIGH | phases/11-llm-engineering/07-advanced-rag/code/main.py | 0 | acme corp security practices. acme corp maintains soc 2 type ii compliance and undergoes annual third-party security aud | STRING |
| HIGH | phases/11-llm-engineering/06-rag/code/main.py | 0 | acme corp security practices. acme corp maintains soc 2 type ii compliance and undergoes annual third-party security aud | STRING |
| HIGH | …es/11-llm-engineering/04-embeddings/code/embeddings.py | 0 | acme corp security practices. acme corp maintains soc 2 type ii compliance and undergoes annual third-party security aud | STRING |
| HIGH | phases/11-llm-engineering/07-advanced-rag/code/main.py | 0 | acme corp api documentation. the acme api uses rest with json request and response bodies. authentication is via bearer | STRING |
| HIGH | phases/11-llm-engineering/06-rag/code/main.py | 0 | acme corp api documentation. the acme api uses rest with json request and response bodies. authentication is via bearer | STRING |
| HIGH | …es/11-llm-engineering/04-embeddings/code/embeddings.py | 0 | acme corp api documentation. the acme api uses rest with json request and response bodies. authentication is via bearer | STRING |
| HIGH | phases/11-llm-engineering/07-advanced-rag/code/main.py | 0 | acme corp uptime and reliability. acme corp guarantees 99.9% uptime for professional plans and 99.99% uptime for enterpr | STRING |
| HIGH | phases/11-llm-engineering/06-rag/code/main.py | 0 | acme corp uptime and reliability. acme corp guarantees 99.9% uptime for professional plans and 99.99% uptime for enterpr | STRING |
| HIGH | …es/11-llm-engineering/04-embeddings/code/embeddings.py | 0 | acme corp uptime and reliability. acme corp guarantees 99.9% uptime for professional plans and 99.99% uptime for enterpr | STRING |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | projects/prompt-regression-tester/solution/main.py | 50 | except (ValueError, TypeError): | CODE |
| HIGH | …t/stages/02-apply-exact-patches/tests/test_stage_02.py | 33 | except ValueError: | CODE |
| HIGH | manuals/docker-sandboxes-101/capture/run.py | 1260 | except ValueError: | CODE |
| HIGH | manuals/docker-sandboxes-101/capture/run.py | 1276 | except ValueError: | CODE |
| HIGH | …sons/11-mcp-server-design-and-integration/code/main.py | 157 | except ValueError: | CODE |
| HIGH | …s/06-speech-and-audio/09-music-generation/code/main.py | 70 | except ValueError: | CODE |
| HIGH | …/11-llm-engineering/03-structured-outputs/code/main.py | 130 | except json.JSONDecodeError: | CODE |
| HIGH | …y/outputs/skill-release-gate/scripts/evaluate_skill.py | 207 | except ValueError: | CODE |
| HIGH | …ent-engineering/37-runtime-feedback-loops/code/main.py | 99 | except FileNotFoundError: | CODE |
| HIGH | …/22-production-scaling-queues-checkpoints/code/main.py | 168 | except RuntimeError: | CODE |
| HIGH | …oundations/11-singular-value-decomposition/code/svd.py | 584 | except ImportError: | CODE |
| HIGH | …19-capstone-projects/52-experiment-runner/code/main.py | 142 | except OSError: | CODE |
| HIGH | …19-capstone-projects/52-experiment-runner/code/main.py | 223 | except OSError: | CODE |
| HIGH | …projects/15-constitutional-safety-harness/code/main.py | 33 | except Exception: | CODE |
| HIGH | …stone-projects/42-large-corpus-downloader/code/main.py | 355 | except Exception: | CODE |
| HIGH | …stone-projects/78-zero-parameter-sharding/code/main.py | 237 | except FileNotFoundError: | CODE |
| HIGH | …stone-projects/78-zero-parameter-sharding/code/main.py | 241 | except OSError: | CODE |
| HIGH | …19-capstone-projects/79-pipeline-parallel/code/main.py | 193 | except FileNotFoundError: | CODE |
| HIGH | …19-capstone-projects/79-pipeline-parallel/code/main.py | 197 | except OSError: | CODE |
| HIGH | …apstone-projects/43-hdf5-tokenized-corpus/code/main.py | 292 | except Exception: | CODE |
| HIGH | …apstone-projects/43-hdf5-tokenized-corpus/code/main.py | 277 | except Exception: | CODE |
| HIGH | …one-projects/28-observability-otel-traces/code/main.py | 474 | except PermissionError: | CODE |
| HIGH | …pstone-projects/47-checkpoint-save-resume/code/main.py | 127 | except FileNotFoundError: | CODE |
| HIGH | …pstone-projects/47-checkpoint-save-resume/code/main.py | 152 | except FileNotFoundError: | CODE |
| HIGH | …one-projects/80-checkpoint-sharded-resume/code/main.py | 85 | except OSError: | CODE |
| HIGH | …e-projects/76-collective-ops-from-scratch/code/main.py | 273 | except FileNotFoundError: | CODE |
| HIGH | …e-projects/76-collective-ops-from-scratch/code/main.py | 277 | except OSError: | CODE |
| HIGH | …19-capstone-projects/77-data-parallel-ddp/code/main.py | 165 | except FileNotFoundError: | CODE |
| HIGH | …19-capstone-projects/77-data-parallel-ddp/code/main.py | 169 | except OSError: | CODE |
| HIGH | …one-projects/83-prompt-injection-detector/code/main.py | 87 | except Exception: | CODE |
| HIGH | …one-projects/83-prompt-injection-detector/code/main.py | 102 | except Exception: | CODE |
| HIGH | …/19-capstone-projects/72-code-exec-metric/code/main.py | 209 | except subprocess.TimeoutExpired: | CODE |
| HIGH | …cts/86-constitutional-rules-engine/code/yaml_subset.py | 70 | except ValueError: | CODE |
| HIGH | …cts/86-constitutional-rules-engine/code/yaml_subset.py | 74 | except ValueError: | CODE |
| HIGH | …l-fundamentals/12-hyperparameter-tuning/code/tuning.py | 560 | except Exception: | CODE |
| HIGH | …ltimodal-ai/09-qwen-vl-family-dynamic-fps/code/main.py | 102 | except json.JSONDecodeError: | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …-budget-planner/stages/01-estimate-cost/starter/cli.py | 35 | except Exception as error: | CODE |
| LOW | projects/agent-budget-planner/solution/cli.py | 35 | except Exception as error: | CODE |
| LOW | …uals/docker-sandboxes-101/research/sources/docs-mcp.md | 934 | except Exception as e: | CODE |
| MEDIUM | …uals/docker-sandboxes-101/research/sources/docs-mcp.md | 935 | print(f"Error: {e}") | CODE |
| LOW | …uals/docker-sandboxes-101/research/sources/docs-mcp.md | 1126 | except Exception as e: | CODE |
| MEDIUM | …uals/docker-sandboxes-101/research/sources/docs-mcp.md | 1127 | print(f"Error: {e}") | CODE |
| LOW | manuals/docker-sandboxes-101/capture/run.py | 1858 | except Exception as exc: | CODE |
| LOW | scripts/link_check.py | 258 | except Exception as exc: | CODE |
| LOW | scripts/link_check.py | 343 | except Exception as exc: | CODE |
| LOW | scripts/check_mcpa_wire.py | 276 | except Exception as exc: | CODE |
| LOW | scripts/check_mcpa_wire.py | 290 | except Exception as exc: | CODE |
| LOW | scripts/install_skills.py | 424 | except Exception: | CODE |
| LOW | scripts/install_skills.py | 525 | except Exception: | CODE |
| LOW | …evals-testing-debugging-and-observability/code/main.py | 62 | except Exception as exc: | CODE |
| LOW | …ude/lessons/10-tool-use-and-agentic-loops/code/main.py | 217 | except Exception as exc: | CODE |
| LOW | …sons/11-mcp-server-design-and-integration/code/main.py | 183 | except Exception: | CODE |
| LOW | …08-messages-api-and-application-lifecycle/code/main.py | 137 | except Exception as exc: # Tool failures become model-visible results. | CODE |
| LOW | …multi-round-trip-requests-and-elicitation/code/main.py | 113 | except Exception: | CODE |
| LOW | …mcpa/lessons/17-tool-invocation-lifecycle/code/main.py | 131 | except Exception: | CODE |
| LOW | …s/mcpa/lessons/33-mcpa-capstone-readiness/code/main.py | 143 | except Exception: | CODE |
| LOW | …pa/lessons/25-consent-and-least-privilege/code/main.py | 70 | except Exception: | CODE |
| LOW | …pa/lessons/25-consent-and-least-privilege/code/main.py | 77 | except Exception: | CODE |
| LOW | …llm-engineering/14-model-context-protocol/code/main.py | 204 | except Exception: | CODE |
| LOW | …llm-engineering/14-model-context-protocol/code/main.py | 232 | except Exception: | CODE |
| LOW | …llm-engineering/14-model-context-protocol/code/main.py | 275 | except Exception: | CODE |
| LOW | …m-engineering/16-langgraph-state-machines/code/main.py | 57 | except Exception as exc: | CODE |
| LOW | …gineering/09-function-calling/code/function_calling.py | 35 | except Exception as e: | CODE |
| LOW | …gineering/09-function-calling/code/function_calling.py | 141 | except Exception as e: | CODE |
| LOW | …ases/11-llm-engineering/09-function-calling/docs/en.md | 221 | except Exception as e: | CODE |
| LOW | …ases/11-llm-engineering/09-function-calling/docs/en.md | 311 | except Exception as e: | CODE |
| LOW | …engineering/02-few-shot-cot/code/advanced_prompting.py | 326 | except Exception as e: | CODE |
| LOW | …-permissions-sandboxes-and-trust/code/sandbox/probe.py | 15 | except Exception as error: | CODE |
| LOW | …s-and-protocols/07-building-an-mcp-server/code/main.py | 355 | except Exception as exc: | CODE |
| LOW | …ools-and-protocols/18-mcp-auth-production/code/main.py | 488 | except Exception: | CODE |
| LOW | …ent-engineering/02-rewoo-plan-and-execute/code/main.py | 43 | except Exception as e: | CODE |
| LOW | …t-engineering/42-agent-workbench-capstone/code/main.py | 547 | except Exception as exc: | CODE |
| LOW | …tputs/agent-workbench-pack/scripts/generate_handoff.py | 112 | except Exception as exc: | CODE |
| LOW | …nt-engineering/10-skill-libraries-voyager/code/main.py | 100 | except Exception as e: | CODE |
| LOW | …/14-agent-engineering/17-claude-agent-sdk/code/main.py | 100 | except Exception as e: | CODE |
| LOW | …gent-engineering/34-repo-memory-and-state/code/main.py | 108 | except Exception: | CODE |
| LOW | …ineering/06-tool-use-and-function-calling/code/main.py | 134 | except Exception as e: | CODE |
| LOW | …-agent-engineering/14-autogen-actor-model/code/main.py | 67 | except Exception as e: | CODE |
| LOW | …es/14-agent-engineering/01-the-agent-loop/code/main.py | 52 | except Exception as e: | CODE |
| LOW | …es/14-agent-engineering/01-the-agent-loop/code/main.py | 62 | except Exception as e: | CODE |
| LOW | …ineering/07-memory-virtual-context-memgpt/code/main.py | 136 | except Exception as e: | CODE |
| LOW | …00-setup-and-tooling/01-dev-environment/code/verify.py | 94 | except Exception as exc: | CODE |
| LOW | …i-agent-and-swarms/08-role-specialization/code/main.py | 75 | except Exception as e: | CODE |
| LOW | …i-agent-and-swarms/08-role-specialization/code/main.py | 84 | except Exception as e: | CODE |
| MEDIUM | …gent-and-swarms/07-society-of-mind-debate/code/main.py | 94 | print(f"Error vs truth: {abs(control_mean - TRUE_ANSWER):.2f}") | CODE |
| LOW | …tch/09-constitutional-ai-self-improvement/code/main.py | 80 | except Exception: | CODE |
| LOW | …tch/09-constitutional-ai-self-improvement/code/main.py | 137 | except Exception: | CODE |
| LOW | …ratch/09-constitutional-ai-self-improvement/docs/en.md | 220 | except Exception: | CODE |
| LOW | …ojects/60-projection-layer-modality-align/code/main.py | 36 | except Exception: | CODE |
| LOW⚡ | …projects/15-constitutional-safety-harness/code/main.py | 33 | except Exception: | CODE |
| LOW | …tone-projects/13-mcp-server-with-registry/code/main.py | 288 | except Exception as exc: | CODE |
| LOW | …stone-projects/42-large-corpus-downloader/code/main.py | 355 | except Exception: | CODE |
| LOW | …apstone-projects/43-hdf5-tokenized-corpus/code/main.py | 273 | except Exception: | CODE |
| LOW | …apstone-projects/43-hdf5-tokenized-corpus/code/main.py | 277 | except Exception: | CODE |
| LOW | …apstone-projects/43-hdf5-tokenized-corpus/code/main.py | 292 | except Exception: | CODE |
| MEDIUM | …apstone-projects/43-hdf5-tokenized-corpus/code/main.py | 140 | def __exit__(self, exc_type, exc, tb) -> None: | CODE |
| 23 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| CRITICAL | scripts/test_agent_api.js | 131 | assert.ok(read.json.result.structuredContent.markdown.startsWith('#')); | CODE |
| CRITICAL | …s/19-capstone-projects/59-vit-transformer/code/main.py | 226 | grad_norm = enc2.front.patch.proj.weight.grad.norm().item() | CODE |
| CRITICAL | …9-capstone-projects/36-training-loop-eval/code/main.py | 159 | block.attn.out_proj.weight.data.mul_(scale) | CODE |
| CRITICAL | …9-capstone-projects/36-training-loop-eval/code/main.py | 160 | block.mlp.fc2.weight.data.mul_(scale) | CODE |
| CRITICAL⚡ | …9-capstone-projects/35-gpt-model-assembly/code/main.py | 156 | block.attn.out_proj.weight.data.mul_(scale) | CODE |
| CRITICAL⚡ | …9-capstone-projects/35-gpt-model-assembly/code/main.py | 157 | block.mlp.fc2.weight.data.mul_(scale) | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | AGENTS.md | 301 | ## New-lesson onboarding | COMMENT |
| LOW | …port-agent/stages/01-search-the-corpus/tests/search.rs | 1 | #[allow(dead_code)] | COMMENT |
| LOW | .github/workflows/translate.yml | 1 | name: translate | COMMENT |
| LOW | …ases/11-llm-engineering/09-function-calling/docs/en.md | 581 | # | COMMENT |
| LOW | …ases/11-llm-engineering/09-function-calling/docs/en.md | 601 | # tools=tools, | COMMENT |
| LOW | …ases/11-llm-engineering/09-function-calling/docs/en.md | 621 | COMMENT | |
| LOW | …ases/11-llm-engineering/09-function-calling/docs/en.md | 641 | # "required": ["city"] | COMMENT |
| LOW | …ases/11-llm-engineering/09-function-calling/docs/en.md | 661 | COMMENT | |
| LOW | phases/11-llm-engineering/10-evaluation/docs/en.md | 741 | # promptfooconfig.yaml: | COMMENT |
| LOW | phases/11-llm-engineering/10-evaluation/docs/en.md | 761 | # | COMMENT |
| LOW | phases/11-llm-engineering/10-evaluation/docs/en.md | 781 | # | COMMENT |
| LOW | phases/11-llm-engineering/10-evaluation/docs/en.md | 801 | # | COMMENT |
| LOW | phases/11-llm-engineering/12-guardrails/docs/en.md | 761 | # model="omni-moderation-latest", | COMMENT |
| LOW | phases/11-llm-engineering/12-guardrails/docs/en.md | 781 | # from transformers import AutoTokenizer, AutoModelForCausalLM | COMMENT |
| LOW | phases/11-llm-engineering/12-guardrails/docs/en.md | 801 | # NeMo Guardrails uses Colang -- a DSL for defining conversational rails. | COMMENT |
| LOW | phases/11-llm-engineering/12-guardrails/docs/en.md | 821 | # user ask about banking | COMMENT |
| LOW | phases/11-llm-engineering/12-guardrails/docs/en.md | 841 | # guard = gd.Guard().use_many( | COMMENT |
| LOW | …es/11-llm-engineering/03-structured-outputs/docs/en.md | 421 | print(f" Output: {json.dumps(result)}") | COMMENT |
| LOW | …es/11-llm-engineering/03-structured-outputs/docs/en.md | 441 | # response = client.beta.chat.completions.parse( | COMMENT |
| LOW | …es/11-llm-engineering/03-structured-outputs/docs/en.md | 461 | # client = anthropic.Anthropic() | COMMENT |
| LOW | …es/11-llm-engineering/03-structured-outputs/docs/en.md | 481 | ``` | COMMENT |
| LOW | phases/11-llm-engineering/13-production-app/docs/en.md | 1021 | ### FastAPI Server (Production Deployment) | COMMENT |
| LOW | phases/11-llm-engineering/13-production-app/docs/en.md | 1041 | # stream: bool = False | COMMENT |
| LOW | phases/11-llm-engineering/13-production-app/docs/en.md | 1061 | # @app.get("/v1/costs") | COMMENT |
| LOW | phases/11-llm-engineering/13-production-app/docs/en.md | 1081 | ```python | COMMENT |
| LOW | phases/11-llm-engineering/13-production-app/docs/en.md | 1101 | # async with client.messages.stream( | COMMENT |
| LOW | …es/11-llm-engineering/01-prompt-engineering/docs/en.md | 881 | COMMENT | |
| LOW | …es/11-llm-engineering/01-prompt-engineering/docs/en.md | 901 | # | COMMENT |
| LOW | …es/11-llm-engineering/01-prompt-engineering/docs/en.md | 921 | # "role": "user", | COMMENT |
| LOW | …es/11-llm-engineering/01-prompt-engineering/docs/en.md | 941 | # from google.genai import types | COMMENT |
| LOW | …es/11-llm-engineering/01-prompt-engineering/docs/en.md | 961 | ```python | COMMENT |
| LOW | phases/11-llm-engineering/11-caching-cost/docs/en.md | 761 | # | COMMENT |
| LOW | phases/11-llm-engineering/11-caching-cost/docs/en.md | 781 | COMMENT | |
| LOW | phases/11-llm-engineering/11-caching-cost/docs/en.md | 801 | # print(f"Completion tokens: {response.usage.completion_tokens}") | COMMENT |
| LOW | phases/11-llm-engineering/11-caching-cost/docs/en.md | 821 | # "model": "gpt-4o-mini", | COMMENT |
| LOW | phases/11-llm-engineering/11-caching-cost/docs/en.md | 841 | # import numpy as np | COMMENT |
| LOW | …tools-and-protocols/01-the-tool-interface/code/main.ts | 1 | // Phase 13 Lesson 01 — the tool interface, in TypeScript. | COMMENT |
| LOW | …es/13-tools-and-protocols/19-a2a-protocol/code/main.ts | 1 | // Phase 13 Lesson 19 — A2A agent-to-agent protocol, in TypeScript. | COMMENT |
| LOW | …-engineering/13-langgraph-stateful-graphs/code/main.ts | 1 | // Phase 14 Lesson 13 — LangGraph-shaped stateful graph, in TypeScript. | COMMENT |
| LOW | …ineering/06-tool-use-and-function-calling/code/main.ts | 1 | // Phase 14 Lesson 06 — tool use and function calling, in TypeScript. | COMMENT |
| LOW | …es/14-agent-engineering/01-the-agent-loop/code/main.ts | 1 | // Phase 14 Lesson 01 — toy ReAct agent loop, in TypeScript. | COMMENT |
| LOW | …/00-setup-and-tooling/10-terminal-and-shell/docs/en.md | 241 | COMMENT | |
| LOW | phases/10-llms-from-scratch/11-quantization/docs/en.md | 781 | # | COMMENT |
| LOW | phases/10-llms-from-scratch/11-quantization/docs/en.md | 801 | # tokenizer = AutoTokenizer.from_pretrained(model_id) | COMMENT |
| LOW | phases/10-llms-from-scratch/11-quantization/docs/en.md | 821 | ### Converting to GGUF | COMMENT |
| LOW | phases/10-llms-from-scratch/10-evaluation/docs/en.md | 421 | The standard tool for running benchmarks on any model. | COMMENT |
| LOW | …jects/03-realtime-voice-assistant/code/ts/src/index.ts | 1 | // Capstone 19/03: realtime voice web client (multi-file TypeScript). | COMMENT |
| LOW | …s/01-terminal-native-coding-agent/code/ts/src/index.ts | 1 | // Capstone 19/01: terminal-native coding agent harness (multi-file TypeScript). | COMMENT |
| LOW | …one-projects/02-rag-over-codebase/code/ts/src/index.ts | 1 | // Capstone 19/02: code RAG query API (multi-file TypeScript). | COMMENT |
| LOW | …utonomous-systems/16-checkpoints-rollback/code/main.py | 101 | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | …/lessons/13-application-security-and-secrets/quiz.json | 22 | "Inside encrypted tool arguments generated by Claude for each authorized request", | CODE |
| HIGH | …llms-from-scratch/06-instruction-tuning-sft/docs/en.md | 25 | Stanford Alpaca proved you don't need millions of examples. In March 2023, they fine-tuned Llama 7B on just 52,000 instr | CODE |
| HIGH | …llms-from-scratch/06-instruction-tuning-sft/docs/en.md | 61 | Simple and widely used. The `input` field is optional -- many instructions don't need additional context. Stanford relea | CODE |
| HIGH | …modal-ai/05-llava-visual-instruction-tuning/docs/en.md | 3 | > LLaVA (April 2023) is the most copied multimodal architecture on the planet. It replaced BLIP-2's Q-Former with a 2-la | CODE |
| HIGH | …modal-ai/05-llava-visual-instruction-tuning/docs/en.md | 168 | | ShareGPT4V | "Better captions" | 1M dense captions generated by GPT-4V, used for higher-quality alignment | | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …s/mcpa/lessons/33-mcpa-capstone-readiness/code/main.py | 635 | CODE | |
| LOW | …oundations/05-chain-rule-and-autodiff/code/autodiff.py | 210 | CODE | |
| LOW | …-math-foundations/21-graph-theory/code/graph_theory.py | 150 | CODE | |
| LOW | …-llms-from-scratch/05-scaling-distributed/code/main.py | 85 | CODE | |
| LOW | …pstone-projects/58-vision-encoder-patches/code/main.py | 102 | CODE | |
| LOW | …-capstone-projects/56-iteration-scheduler/code/main.py | 117 | CODE | |
| LOW | …19-capstone-projects/34-transformer-block/code/main.py | 111 | CODE | |
| LOW | …pstone-projects/47-checkpoint-save-resume/code/main.py | 374 | CODE | |
| LOW | …s/19-capstone-projects/59-vit-transformer/code/main.py | 103 | CODE | |
| LOW | …capstone-projects/48-distributed-fsdp-ddp/code/main.py | 294 | CODE | |
| LOW | …/85-content-classifier-integration/code/classifiers.py | 148 | CODE | |
| LOW | …9-capstone-projects/35-gpt-model-assembly/code/main.py | 91 | CODE | |
| LOW | …04-computer-vision/14-vision-transformers/code/main.py | 40 | CODE | |
| LOW | …odal-ai/04-flamingo-gated-cross-attention/code/main.py | 86 | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM⚡ | site/glossary.html | 26 | <script type="application/ld+json">{"@context":"https://schema.org","@type":"DefinedTermSet","@id":"https://aiengineer | CODE |
| MEDIUM⚡ | …ne-projects/27-eval-harness-fixture-tasks/code/main.py | 332 | # The harness | COMMENT |
| MEDIUM⚡ | …e-projects/29-end-to-end-coding-task-demo/code/main.py | 32 | # Minimal harness primitives, copied with intent from lessons 25-28. | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | scripts/scaffold-lesson.sh | 77 | ### Step 1: [name] | COMMENT |
| LOW | scripts/scaffold-lesson.sh | 85 | ### Step 2: [name] | COMMENT |
| LOW⚡ | …s/15-autonomous-systems/11-browser-agents/code/main.py | 91 | # Step 1: fetch. Apply sanitizer if enabled. | COMMENT |
| LOW⚡ | …s/15-autonomous-systems/11-browser-agents/code/main.py | 97 | # Step 2: "read" the combined context. In a real stack, URL fragment | COMMENT |
| LOW⚡ | …s/15-autonomous-systems/11-browser-agents/code/main.py | 101 | # Step 3: decide what to post. | COMMENT |
| LOW | …s/15-autonomous-systems/11-browser-agents/code/main.py | 117 | # Step 4: R/W boundary check. | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | …1-llm-engineering/10-evaluation/code/eval_framework.py | 259 | reference_output="I cannot provide instructions for illegal activities.", | CODE |
| MEDIUM | phases/11-llm-engineering/12-guardrails/code/main.ts | 327 | return { response: "I cannot provide that response. Let me help you differently.", report }; | CODE |
| MEDIUM | …es/11-llm-engineering/12-guardrails/code/guardrails.py | 341 | return "I apologize, but I cannot provide that response. Let me help you differently.", report | CODE |
| MEDIUM | …m-engineering/13-production-app/code/production_app.py | 352 | "Let me know if you need more detail on any specific aspect." | CODE |
| MEDIUM | …m-engineering/13-production-app/code/production_app.py | 493 | result["text"] = "I cannot provide that response as it was flagged by our safety system." | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …ocker-sandboxes-101/research/sources/docs-sandboxes.md | 9914 | "org_name": "Acme Inc", | CODE |
| LOW | …ocker-sandboxes-101/research/sources/docs-sandboxes.md | 10118 | "org_name": "Acme Inc", | CODE |
| LOW | …ocker-sandboxes-101/research/sources/docs-sandboxes.md | 10147 | "org_name": "Acme Inc", | CODE |
| LOW | manuals/docker-sandboxes-101/capture/run.py | 680 | return [["sbx", "secret", "rm", "--placeholder", secret["placeholder"], "-f"] for secret in kit.custom_secrets() if | CODE |
| LOW | manuals/docker-sandboxes-101/capture/run.py | 902 | placeholder = next((secret["placeholder"] for secret in kit.custom_secrets() if secret.get("env") == "M101_RECV_ | CODE |
| LOW | site/ui-i18n.js | 5 | var ATTRS = ['aria-label', 'title', 'placeholder']; | CODE |
| LOW | …-tools-and-protocols/04-structured-output/code/main.py | 143 | "customer": "Acme Corp", | CODE |
| LOW | …dal-ai/05-llava-visual-instruction-tuning/code/main.py | 76 | "placeholder": placeholder, | CODE |
| LOW | …utonomous-systems/14-kill-switches-canaries/docs/en.md | 97 | 1. Run `code/main.py`. Confirm the circuit breaker fires on turn 5 (fifth identical call) and the canary fires on turn 9 | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …-engineering/13-langgraph-stateful-graphs/code/main.py | 41 | def set_entry(self, name: str) -> None: | CODE |
| LOW | …tooling/12-debugging-and-profiling/code/debug_tools.py | 10 | logger = logging.getLogger(__name__) | CODE |
| LOW | …nd-swarms/17-generative-agents-simulation/code/main.py | 53 | def update_plan(self, tick: int) -> None: | CODE |
| LOW | …pstone-projects/66-reranker-cross-encoder/code/main.py | 35 | def _set_seed(seed: int = SEED) -> None: | CODE |
| LOW | …19-capstone-projects/34-transformer-block/code/main.py | 195 | def _set_eval_mode(stack: BlockStack) -> None: | CODE |
| LOW | …apstone-projects/45-gradient-clipping-amp/code/main.py | 183 | def set_loss_fn(self, fn: Callable[[torch.Tensor, torch.Tensor], torch.Tensor]) -> None: | CODE |
| LOW | …apstone-projects/45-gradient-clipping-amp/code/main.py | 186 | def set_lr(self, lr: float) -> None: | CODE |
| LOW | …/19-capstone-projects/44-cosine-lr-warmup/code/main.py | 150 | def set_lr(self, lr: float) -> None: | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | …cations/mcpa/lessons/03-json-rpc-and-meta/code/main.py | 146 | def handle_request(message: dict) -> dict: | CODE |
| LOW | …m-engineering/13-production-app/code/production_app.py | 445 | async def handle_request(self, user_id, query, template_name="general_chat", variables=None): | CODE |
| LOW | …t-engineering/19-benchmarks-swebench-gaia/code/main.py | 34 | def run_task(task: Task) -> TaskResult: | CODE |
| LOW | …rms/24-evaluation-coordination-benchmarks/code/main.py | 42 | def run_task(system: SystemSim, task_id: str, seen: bool, rng: random.Random) -> TaskResult: | CODE |
| LOW | …s/19-capstone-projects/49-lm-eval-harness/code/main.py | 289 | def run_task( | CODE |
| LOW | …odal-ai/25-multimodal-agents-computer-use/code/main.py | 87 | def run_task(task: Task) -> dict: | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | …engineering/02-few-shot-cot/code/advanced_prompting.py | 508 | client = OpenAI(api_key=os.environ.get("OPENAI_API_KEY", "your-api-key")) | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | …oud-agent-with-aws-strands/solution/strands_adapter.py | 33 | CODE |