Home / JuliusBrussee / caveman
AI Slop Code RIsk Monitor

JuliusBrussee/caveman

🪨 why use many token when few token do trick. Viral skill + proxy for coding agents that cuts 65% of tokens by talking like a caveman.

7.2 Moderate pattern signal View on GitHub

Maintenance and Governance Review

This report summarizes scanner evidence for JuliusBrussee/caveman, a Go project with 110,916 GitHub stars. SynthScan examined 292,904 lines of code across 1161 source files, recording 1572 pattern matches distributed across 21 syntactic categories. The legacy adjusted pattern signal is 7.2 (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.

7.2
Adjusted Signal
7.2
Raw Pattern Score
100%
Time Factor
2026-10-10
Last Push
110.9K
Stars
Go
Language
292.9K
Lines of Code
1.2K
Files
1.6K
Pattern Hits
2026-10-12
Scan Date
0.08
HC Hit Rate

Risk Review · v3

Use each vector to guide manual verification of the underlying findings.

Cognitive Ownership Risk 17.1/100 · LOW

Can maintainers explain intent, assumptions, and trade-offs? 1126 finding(s).

Hallucinated Dependency & Supply Chain Risk 5.6/100 · LOW

Are imports declared and package/API names verified? 166 finding(s).

Agent Over-Engineering Ratio 19.6/100 · LOW

Are wrappers, guards, and repeated structures necessary? 951 finding(s).

PR Review Velocity Impact 33.0/100 · MEDIUM

Which changes need focused human review? 1572 finding(s).

What These Metrics Mean

Adjusted Signal
Legacy pattern score normalized per 1,000 scanned lines and multiplied by the repository-level time factor. It is not an estimate of AI authorship.
Raw Pattern Score
Sum of severity- and context-weighted pattern findings before applying the time factor.
Time Factor
A coarse multiplier based on repository last-push metadata. It does not identify when individual code was written.
Pattern Hits
Count of configured pattern findings. Review each location and surrounding code before acting.
HC Hit Rate
High- and critical-severity findings per scanned file. Severity is a configured detector weight, not proof of a vulnerability or authorship.
Lines / Files
Scan coverage denominators used by the legacy normalized score.

Pattern-Signal History

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.

Severity Breakdown

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.

CRITICAL 3HIGH 85MEDIUM 163LOW 1321

Directory Score Breakdown

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.

Pattern Findings

The scanner identified 1572 distinct pattern matches across 21 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.

Hyper-Verbose Identifiers668 hits · 596 pts
SeverityFileLineSnippetContext
LOW…ns/caveman/skills/caveman-compress/scripts/validate.py121def extract_indented_code_blocks(text):CODE
LOW…ns/caveman/skills/caveman-compress/scripts/compress.py420def resolve_max_output_tokens() -> int:CODE
LOWtests/test_eval_fidelity.py105 def test_case_set_mismatch_is_refused(self) -> None:CODE
LOWtests/test_eval_fidelity.py124 def test_cases_are_well_formed(self) -> None:CODE
LOWtests/test_eval_fidelity.py145 def test_examples_match_checks(self) -> None:CODE
LOWtests/test_eval_fidelity.py168 def test_every_category_covered(self) -> None:CODE
LOWtests/test_eval_llm_run.py139 def test_usage_is_stored_alongside_text(self) -> None:CODE
LOWtests/test_eval_llm_run.py152 def test_error_result_is_a_failure(self) -> None:CODE
LOWtests/test_eval_llm_run.py165 def test_transient_failure_is_retried(self) -> None:CODE
LOWtests/test_eval_llm_run.py176 def test_hard_failure_writes_partial_and_exits(self) -> None:CODE
LOWtests/test_eval_llm_run.py198 def test_language_selects_control_and_snapshot(self) -> None:CODE
LOWtests/test_eval_llm_run.py210 def test_fidelity_set_runs_case_prompts(self) -> None:CODE
LOWtests/test_eval_llm_run.py231 def test_bad_eval_set_exits_before_any_call(self) -> None:CODE
LOWtests/test_eval_llm_run.py244 def test_reminder_run_ignores_eval_set_and_lang(self) -> None:CODE
LOWtests/test_eval_llm_run.py254 def test_unknown_language_exits_before_any_call(self) -> None:CODE
LOWtests/test_compress_cli.py25 def test_help_and_argument_errors_for_direct_and_module_entrypoints(self):CODE
LOWtests/test_compress_cli.py35 def test_relative_code_path_is_skipped_without_writing_or_spawning(self):CODE
LOWtests/test_compress_cli.py46 def test_missing_unicode_file_reports_error_without_import_or_encoding_failure(self):CODE
LOWtests/verify_repo.py121def verify_shipped_skills_are_documented() -> None:CODE
LOWtests/verify_repo.py162def verify_skills_root_holds_only_skills() -> None:CODE
LOWtests/verify_repo.py205def verify_skill_frontmatter_upload_compatibility() -> None:CODE
LOWtests/verify_repo.py302def verify_manifests_and_syntax() -> None:CODE
LOWtests/verify_repo.py557def verify_python_text_io_encoding() -> None:CODE
LOWtests/verify_repo.py904def verify_license_boundaries() -> None:CODE
LOWtests/verify_repo.py1076def verify_untrusted_git_invocations() -> None:CODE
LOW⚡tests/test_detect.py68 def test_known_names_case_insensitive(self):STRING
LOW⚡tests/test_detect.py74 def test_cmakelists_txt_not_compressible_despite_txt_extension(self):STRING
LOW⚡tests/test_detect.py80 def test_shebang_script_is_code(self):STRING
LOW⚡tests/test_detect.py86 def test_extensionless_prose_still_compressible(self):STRING
LOW⚡tests/test_detect.py92 def test_markdown_still_compressible(self):STRING
LOW⚡tests/test_detect.py98 def test_cursor_rule_mdc_is_natural_language(self):STRING
LOW⚡tests/test_mode_tracker.py98 def test_turn_caveman_mode_off_deactivates(self):CODE
LOW⚡tests/test_mode_tracker.py104 def test_turn_caveman_off_deactivates(self):CODE
LOW⚡tests/test_mode_tracker.py109 def test_turn_off_caveman_deactivates(self):CODE
LOW⚡tests/test_mode_tracker.py114 def test_stop_caveman_multiline_deactivates(self):CODE
LOW⚡tests/test_mode_tracker.py120 def test_normal_mode_command_deactivates(self):CODE
LOW⚡tests/test_mode_tracker.py125 def test_back_to_normal_mode_deactivates(self):CODE
LOW⚡tests/test_mode_tracker.py130 def test_vim_normal_mode_does_not_deactivate(self):CODE
LOW⚡tests/test_mode_tracker.py137 def test_enable_caveman_with_stop_elsewhere_activates(self):CODE
LOW⚡tests/test_mode_tracker.py144 def test_question_does_not_activate(self):CODE
LOW⚡tests/test_mode_tracker.py150 def test_scoped_brevity_does_not_activate(self):CODE
LOW⚡tests/test_mode_tracker.py154 def test_unscoped_brevity_activates(self):CODE
LOW⚡tests/test_mode_tracker.py158 def test_activate_caveman_still_works(self):CODE
LOW⚡tests/test_mode_tracker.py162 def test_turn_on_caveman_mode_still_works(self):CODE
LOW⚡tests/test_mode_tracker.py166 def test_talk_like_caveman_still_works(self):CODE
LOW⚡tests/test_mode_tracker.py170 def test_bare_caveman_mode_still_works(self):CODE
LOW⚡tests/test_mode_tracker.py291 def test_commit_restores_prior_mode_on_next_prompt(self):CODE
LOW⚡tests/test_mode_tracker.py300 def test_commit_with_no_prior_mode_deactivates_after(self):CODE
LOW⚡tests/test_mode_tracker.py307 def test_chained_independent_modes_keep_original_prev(self):CODE
LOW⚡tests/test_mode_tracker.py315 def test_namespaced_commit_and_review_recognized(self):CODE
LOW⚡tests/test_mode_tracker.py324 def test_no_reinforcement_during_independent_turn(self):CODE
LOW⚡tests/test_mode_tracker.py329 def test_deactivation_clears_saved_prev(self):CODE
LOW⚡tests/test_mode_tracker.py436 def test_legacy_mirror_never_holds_literal_off(self):CODE
LOW⚡tests/test_mode_tracker.py445 def test_legacy_mirror_tracks_last_write(self):CODE
LOW⚡tests/test_mode_tracker.py450 def test_prev_restore_is_session_scoped(self):CODE
LOW⚡tests/test_mode_tracker.py462 def test_malformed_session_id_falls_back_to_legacy(self):CODE
LOW⚡tests/test_mode_tracker.py470 def test_session_id_cannot_escape_the_sessions_directory(self):CODE
LOW⚡tests/test_mode_tracker.py476 def test_one_shot_skill_does_not_switch_an_off_session_on(self):CODE
LOWtests/test_mode_tracker.py60 def test_acts_on_first_complete_payload_without_waiting_for_eof(self):CODE
LOWtests/test_mode_tracker.py193 def test_status_reports_active_mode_without_mutating_state(self):CODE
608 more matches not shown…
Over-Commented Block358 hits · 340 pts
SeverityFileLineSnippetContext
LOWinstall.sh1#!/usr/bin/env bashCOMMENT
LOWproxy/doc.go1// Package proxy is the public, byte-safe Caveman reverse proxy: the `caveman`COMMENT
LOWproxy/cmd/caveman-proxy/main.go1// Command caveman-proxy is the standalone byte-safe proxy binary. `caveman start`COMMENT
LOWproxy/providers/toolzone.go1package providersCOMMENT
LOWproxy/providers/adapter.go81 // different, route-scoped env key.COMMENT
LOWproxy/providers/adapter.go201 // seam — so adapter checks on this field are standalone-only guards. EmptyCOMMENT
LOWproxy/providers/adapter.go261 ReasoningTokens intCOMMENT
LOWproxy/providers/adapter.go281 // chunk's null for a key never erases an earlier chunk's non-null value forCOMMENT
LOWproxy/providers/adapter.go301 ObservationCount intCOMMENT
LOWproxy/providers/adapter.go341type Adapter interface {COMMENT
LOWproxy/providers/adapter.go361 ParseUsage(ctx context.Context, responseHeaders http.Header, streamOrBody io.Reader) (UsageObservation, io.Reader, erroCOMMENT
LOWproxy/providers/adapter.go381// metadata spelling with another provider. Bare GET /v1/models is bothCOMMENT
LOWproxy/providers/adapter.go401// RewritableBlock is one content block an adapter exposes for byte-safe rewriting,COMMENT
LOWproxy/providers/adapter.go441 BaseURL stringCOMMENT
LOWproxy/providers/adapter.go1281}COMMENT
LOWproxy/providers/adapter.go1301// ParseUsageBytes a single synthesized {"usage":{...}} document, which carriesCOMMENT
LOWproxy/providers/gemini/content_compress.go81 }COMMENT
LOWproxy/providers/gemini/explicit_cache.go21// ApplyProviderNativeTransforms is an unconditional byte-safe pass-through.COMMENT
LOWproxy/providers/openaicompat/openaicompat.go201// anthropic by construction). The Anthropic grammar is required there, not aCOMMENT
LOWproxy/providers/openaicompat/openaicompat.go241 return inspectOpenAICompatible(ctx, a.Base, body, headers)COMMENT
LOWproxy/providers/openaicompat/openaicompat.go381// different wire grammar than the mount's OpenAI-compatible identity.COMMENT
LOWproxy/providers/bedrock/cache_points.go121// for a model: an Anthropic Claude id — bare or behind one inference-profileCOMMENT
LOWproxy/providers/bedrock/bedrock.go1package bedrockCOMMENT
LOWproxy/providers/bedrock/bedrock.go21// POST /bedrock/model/{modelId}/invokeCOMMENT
LOWproxy/providers/bedrock/usage.go21COMMENT
LOWproxy/providers/anthropic/breakpoint_plan.go21COMMENT
LOWproxy/providers/anthropic/breakpoint_plan.go41// changes no model-visible bytes; on any parse anomaly the body is left unchanged.COMMENT
LOWproxy/providers/anthropic/breakpoint_plan.go241// The documented semantics (platform.claude.com/docs/en/build-with-claude/prompt-caching):COMMENT
LOWproxy/providers/anthropic/cogs_bench_test.go1package anthropicCOMMENT
LOWproxy/providers/vertex/vertex.go21// the gateway forwards it and holds no GCP secret.COMMENT
LOWproxy/providers/openai/cache_key.go21 providers.BaseCOMMENT
LOWproxy/providers/openai/cache_key.go101 // for that request; the mutation is not.COMMENT
LOWproxy/providers/openai/cache_key.go121//COMMENT
LOWproxy/providers/openai/count_tokens.go61// unprojectableResponsesFields name SERVER-HELD prompt surface: content theCOMMENT
LOWproxy/providers/openai/count_tokens.go81//COMMENT
LOWproxy/providers/openai/reasoning_effort.go41//COMMENT
LOWproxy/providers/openai/reasoning_effort.go141 }COMMENT
LOWproxy/providers/openai/session_cache_key.go21// sessionCacheKeyDomain separates this hash from any other sha256 of the sameCOMMENT
LOWproxy/providers/openai/output_brevity.go1package openaiCOMMENT
LOWproxy/providers/openai/stream_usage.go1package openaiCOMMENT
LOWproxy/providers/openai/stream_usage.go21// OpenAI-compatible client already knows to either consume or ignore anCOMMENT
LOWproxy/providers/openai/stream_usage.go61 }COMMENT
LOWproxy/internal/middleware/seal.go21// authority does not open.COMMENT
LOWproxy/internal/middleware/runtime.go41 // report persistent:false and nothing is compressed that could not beCOMMENT
LOWproxy/internal/middleware/runtime.go621}COMMENT
LOWproxy/internal/identity/identity.go1// Package identity resolves who is calling the framework middleware routes andCOMMENT
LOWproxy/internal/standalone/standalone.go41// returns a static context built from caveman.yaml. There is still noCOMMENT
LOWproxy/internal/standalone/standalone.go261 return strings.TrimSpace(strings.TrimPrefix(value, "Bearer "))COMMENT
LOWproxy/internal/standalone/standalone.go281 // MCP tool (over the shared CCR store) instead of the proxy's server-side loop,COMMENT
LOWproxy/internal/standalone/standalone.go521// dial time so a malicious or misconfigured request can't make the local proxyCOMMENT
LOWproxy/internal/standalone/bedrock_chain.go1package standaloneCOMMENT
LOWproxy/internal/config/config.go41 // Listen is the host:port standalone mode binds to.COMMENT
LOWproxy/internal/config/config.go61 // "off" disables it; unknown values fail closed to "off". There isCOMMENT
LOWproxy/internal/config/config.go81 // record mode stays byte-safe pass-through and savings_usd stays 0.COMMENT
LOWproxy/internal/config/config.go101 // Code works unchanged. All bundles are additive; a corrupt one fails Load.COMMENT
LOWproxy/internal/config/config.go161 // wire shape: its usage block reports input_tokens exclusive of cacheCOMMENT
LOWproxy/internal/config/config.go581 return credentialCOMMENT
LOWproxy/internal/config/middleware.go41 QuotaRows int64 `yaml:"quota_rows"` // per principalCOMMENT
LOWproxy/internal/nativeruntime/runtime.go1201COMMENT
LOWproxy/internal/nativeruntime/runtime.go1221// webhook-delivery-gaps served JSON and CSV pages over the threshold, were maskedCOMMENT
298 more matches not shown…
Decorative Section Separators117 hits · 331 pts
SeverityFileLineSnippetContext
MEDIUM⚡evals/snapshots/results.json24 "# Database Connection Pooling\n\n## The problem\n\nOpening a database connection is expensive. One connection usuCODE
MEDIUM⚡evals/snapshots/results.json32 "The main question is how many consumers should get each message.\n\n## Queue (point-to-point)\n**Each message is CODE
MEDIUMextension/popup.js22// ── pixel-flame logo (the CaveMark, 8×8) — white on onyx, no colour ─────────COMMENT
MEDIUMextension/popup.js44// ── settings wiring ──────────────────────────────────────────────────────COMMENT
MEDIUMproxy/providers/gemini/content_compress_test.go56// ── live-zone cost ────────────────────────────────────────────────────────────COMMENT
MEDIUMproxy/providers/openai/content_compress_test.go185// ── live-zone cost ────────────────────────────────────────────────────────────COMMENT
MEDIUMtests/test_caveman_stats.js685// ── statusline: per-session badge ──────────────────────────────────────────COMMENT
MEDIUMtests/test_caveman_stats.js827// ── No counterfactual from transcript usage ────────────────────────────COMMENT
MEDIUMtests/test_caveman_stats.js896// ── Mid-session mode-change attribution (#601) ─────────────────────────────COMMENT
MEDIUMtests/test_caveman_stats.js896// ── Mid-session mode-change attribution (#601) ─────────────────────────────COMMENT
MEDIUMtests/test_caveman_stats.js1037// ── No net or overhead claims without a measured comparison ────────────COMMENT
MEDIUMtests/test_mcp_shrink.js344// ── Packaging (#597) ────────────────────────────────────────────────────────COMMENT
MEDIUMtests/test_mcp_shrink.js344// ── Packaging (#597) ────────────────────────────────────────────────────────COMMENT
MEDIUMtests/test_mcp_shrink.js383// ── Teardown (shutdown.js) ──────────────────────────────────────────────────COMMENT
MEDIUMtests/test_hook_language_rule.js97// ── The two delivery paths ──────────────────────────────────────────────────COMMENT
MEDIUMtests/test_hook_language_rule.js155// ── The same example must not come back anywhere it is injected from ────────COMMENT
MEDIUMtests/test_repo_local_config.js161// ── #634: optional startDir param (backward compatible) ────────────────────COMMENT
MEDIUMtests/test_repo_local_config.js161// ── #634: optional startDir param (backward compatible) ────────────────────COMMENT
MEDIUM⚡tests/test_mode_tracker.py96 # ── #598: deactivation word orders ──────────────────────────────────COMMENT
MEDIUM⚡tests/test_mode_tracker.py135 # ── #598: activation guards ─────────────────────────────────────────COMMENT
MEDIUM⚡tests/test_mode_tracker.py174 # ── slash commands ──────────────────────────────────────────────────COMMENT
MEDIUM⚡tests/test_mode_tracker.py289 # ── #599: one-shot independent modes ────────────────────────────────COMMENT
MEDIUMtests/test_mode_tracker.py58 # ── hook budget: act on the first complete payload, not on EOF ──────COMMENT
MEDIUMtests/test_mode_tracker.py191 # ── #856: read-only mode status ──────────────────────────────────────COMMENT
MEDIUMtests/test_cavecrew_model_overrides.js30// ── patchFrontmatterModel ──────────────────────────────────────────────────COMMENT
MEDIUMtests/test_cavecrew_model_overrides.js155// ── resolvePluginRoot ──────────────────────────────────────────────────────COMMENT
MEDIUMtests/test_cavecrew_model_overrides.js165// ── applyOverrides ─────────────────────────────────────────────────────────COMMENT
MEDIUMtests/test_cavecrew_model_overrides.js283// ── Summary ────────────────────────────────────────────────────────────────COMMENT
MEDIUMtests/test_cavecrew_model_overrides.js286// ── source checkout guard ──────────────────────────────────────────────────COMMENT
MEDIUMshared/platform/ssrf/ssrf_test.go35 // ── Always-blocked IP literals ──────────────────────────────────────────COMMENT
MEDIUMshared/platform/ssrf/ssrf_test.go152 // ── RFC1918 private — blocked in managed mode ────────────────────────────COMMENT
MEDIUMshared/platform/ssrf/ssrf_test.go197 // ── Scheme/credential/port checks ────────────────────────────────────────COMMENT
MEDIUMshared/platform/ssrf/ssrf_test.go230 // ── Localhost hostname ────────────────────────────────────────────────────COMMENT
MEDIUMshared/platform/ssrf/ssrf_test.go245 // ── Public addresses allowed ──────────────────────────────────────────────COMMENT
MEDIUM⚡packages/cli/src/index.ts10887// ── caveman sync ─────────────────────────────────────────────────────────────COMMENT
MEDIUMpackages/cli/src/index.ts2394// ── caveman setup ────────────────────────────────────────────────────────────COMMENT
MEDIUMpackages/cli/src/index.ts2809// ── caveman update ───────────────────────────────────────────────────────────COMMENT
MEDIUMpackages/cli/src/index.ts3860// ── Wrap entitlement: an ACCOUNT fact, never a compression gate ──────COMMENT
MEDIUMpackages/cli/src/index.ts4281// ── entitlement handshake (called at login success) ─────────────────────────────COMMENT
MEDIUMpackages/cli/src/index.ts4474// ── wrap-start + session-end lines (the §8 golden transcripts) ───────────────────COMMENT
MEDIUMpackages/cli/src/index.ts11684// ── caveman mcp install ──────────────────────────────────────────────────────COMMENT
MEDIUMpackages/cli/src/index.ts14662// ── command-output auto-hook (the RTK-parity loadout layer) ──────────────────COMMENT
MEDIUMpackages/cli/src/index.ts15726// ── opt-in auto-recall hook (cavemem) ───────────────────────────────────────COMMENT
MEDIUMpackages/cli/src/index.ts15876// ── soft tier: instruction-note ──────────────────────────────────────────────COMMENT
MEDIUMpackages/cli/src/index.ts16020// ── wrap directives (AUTOPILOT_SPEC §8.4/§7.3, slice D) ─────────────────────COMMENT
MEDIUMpackages/cli/src/index.ts16072// ── hard tier: opencode plugin ───────────────────────────────────────────────COMMENT
MEDIUMpackages/cli/src/index.ts16157// ── hard tier: Hermes plugin ─────────────────────────────────────────────────COMMENT
MEDIUM⚡packages/sdk/python/tests/test_runtime_policy.py163# ─── signature cases ────────────────────────────────────────────────────────COMMENT
MEDIUM⚡packages/sdk/python/tests/test_runtime_policy.py430# ─── kill paths ─────────────────────────────────────────────────────────────COMMENT
MEDIUM⚡packages/sdk/python/tests/test_runtime_policy.py460# ─── refresh robustness ─────────────────────────────────────────────────────COMMENT
MEDIUMpackages/sdk/python/tests/test_runtime_policy.py132# ─── fetch: the wire ────────────────────────────────────────────────────────COMMENT
MEDIUMpackages/sdk/python/tests/test_runtime_policy.py382# ─── decision cases ─────────────────────────────────────────────────────────COMMENT
MEDIUMpackages/sdk/python/tests/test_runtime_policy.py580# ─── decide() is total ──────────────────────────────────────────────────────COMMENT
MEDIUMpackages/sdk/python/tests/test_runtime_policy.py635# ─── fail-closed policy shapes ──────────────────────────────────────────────COMMENT
MEDIUMpackages/sdk/python/tests/test_runtime_policy.py843# ─── guard cases (the shared fail-closed truth table) ───────────────────────COMMENT
MEDIUMpackages/sdk/python/tests/test_runtime_policy.py923# ─── decision span ──────────────────────────────────────────────────────────COMMENT
MEDIUMpackages/sdk/python/tests/test_parity.py74# ─── Per-operation handlers (return the canonical, snake-keyed result) ──────────COMMENT
MEDIUMpackages/sdk/python/tests/test_parity.py371# ─── Drive every fixture ────────────────────────────────────────────────────────COMMENT
MEDIUMpackages/sdk/python/caveman_cloud/core.py2280 # ── lifecycle ────────────────────────────────────────────────────────────COMMENT
MEDIUMpackages/sdk/python/caveman_cloud/core.py2311 # ── refresh ──────────────────────────────────────────────────────────────COMMENT
57 more matches not shown…
Empty Error Handlers43 hits · 215 pts
SeverityFileLineSnippetContext
HIGHevals/measure.py42except Exception:CODE
HIGHevals/plot.py29except Exception:CODE
HIGHevals/score_fidelity.py32except Exception:CODE
HIGHevals/llm_run.py74except Exception:CODE
HIGH.blocks/release-smoke.py119except (OSError, ValueError):CODE
HIGHplugins/caveman/skills/caveman-compress/scripts/cli.py20except Exception:CODE
HIGH…ns/caveman/skills/caveman-compress/scripts/compress.py39except Exception:CODE
HIGH…ns/caveman/skills/caveman-compress/scripts/compress.py148except OSError:CODE
HIGH…ns/caveman/skills/caveman-compress/scripts/compress.py737except AnthropicSdkUnavailable:CODE
HIGH…ns/caveman/skills/caveman-compress/scripts/compress.py1002except OSError:CODE
HIGH…ns/caveman/skills/caveman-compress/scripts/compress.py297except OSError:CODE
HIGHtests/verify_repo.py25except Exception:CODE
HIGHtests/test_mode_tracker.py91except Exception:CODE
HIGHbenchmarks/run.py21except Exception:CODE
HIGH…ckages/middleware/python/caveman_middleware/_httpx2.py42except http.RequestNotRead:CODE
HIGHpackages/middleware/python/caveman_middleware/google.py381except (httpx.RequestNotRead, UnicodeError):CODE
HIGHpackages/middleware/python/caveman_middleware/google.py418except (TypeError, ValueError):CODE
HIGHpackages/middleware/python/caveman_middleware/google.py411except (TypeError, ValueError):CODE
HIGHpackages/middleware/python/caveman_middleware/_usage.py67except (ValueError, TypeError):CODE
HIGHpackages/sdk/python/tests/test_middleware_transport.py108except OSError:CODE
HIGHpackages/sdk/python/tests/test_middleware_transport.py348except TimeoutError:CODE
HIGHpackages/sdk/python/tests/test_middleware_transport.py268except OSError:CODE
HIGHpackages/sdk/python/caveman_cloud/core.py1178except BaseException: # noqa: BLE001 — telemetry must never replace a tool result/error.CODE
HIGHpackages/sdk/python/caveman_cloud/core.py2174except Exception: # noqa: BLE001 — observability must never break a decision.CODE
HIGHpackages/sdk/python/caveman_cloud/core.py2455except Exception: # noqa: BLE001 — observability never breaks a decision.CODE
HIGH…kages/sdk/python/caveman_cloud/middleware/transport.py42except ValueError:CODE
HIGH…kages/sdk/python/caveman_cloud/middleware/transport.py66except OSError:CODE
HIGH…kages/sdk/python/caveman_cloud/middleware/transport.py118except (OSError, ValueError):CODE
HIGH…kages/sdk/python/caveman_cloud/middleware/transport.py216except OSError:CODE
HIGH…ckages/sdk/python/caveman_cloud/middleware/protocol.py252except UnicodeEncodeError:CODE
HIGHpackages/sdk/python/caveman_cloud/middleware/runtime.py163except Exception:CODE
HIGHpackages/sdk/python/caveman_cloud/middleware/runtime.py557except Exception:CODE
HIGHpackages/sdk/python/caveman_cloud/middleware/runtime.py769except Exception:CODE
HIGHpackages/sdk/python/caveman_cloud/middleware/runtime.py431except Exception:CODE
HIGHpackages/sdk/python/caveman_cloud/middleware/runtime.py709except Exception:CODE
HIGHpackages/sdk/python/caveman_cloud/middleware/runtime.py714except Exception:CODE
HIGH…s/sdk/python/caveman_cloud/middleware/async_runtime.py142except MiddlewareError:CODE
HIGHskills/caveman-compress/scripts/cli.py20except Exception:CODE
HIGHskills/caveman-compress/scripts/compress.py39except Exception:CODE
HIGHskills/caveman-compress/scripts/compress.py148except OSError:CODE
HIGHskills/caveman-compress/scripts/compress.py737except AnthropicSdkUnavailable:CODE
HIGHskills/caveman-compress/scripts/compress.py1002except OSError:CODE
HIGHskills/caveman-compress/scripts/compress.py297except OSError:CODE
Cross-Language Confusion (JS/TS)37 hits · 165 pts
SeverityFileLineSnippetContext
HIGH⚡packages/cli/src/index.ts8362 return NoneCODE
HIGH⚡packages/cli/src/index.ts8366 return NoneCODE
HIGH⚡packages/cli/src/index.ts10871 print({ logged_out: true, remote_session_revoked: true, external_token_cleared: false, external_token_source: "CAVE_TCODE
HIGH⚡packages/cli/src/index.ts10880 print({ logged_out: true });CODE
HIGH⚡packages/cli/src/index.ts16212 return FalseCODE
HIGH⚡packages/cli/src/index.ts16223 return FalseCODE
HIGH⚡packages/cli/src/index.ts16229 return NoneCODE
HIGH⚡packages/cli/src/index.ts16232 return NoneCODE
HIGH⚡packages/cli/src/index.ts16234 return NoneCODE
HIGH⚡packages/cli/src/index.ts16236 return NoneCODE
HIGH⚡packages/cli/src/index.ts16245 return NoneCODE
HIGH⚡packages/cli/src/index.ts16248 return NoneCODE
HIGHpackages/cli/src/index.ts764 print({CODE
HIGHpackages/cli/src/index.ts786 print({ telemetry: "on", anonymous_id: anonymousId, note: "env override active (DO_NOT_TRACK/CAVEMAN_TELEMETRY) — noCODE
HIGHpackages/cli/src/index.ts789 print({ telemetry: "on", anonymous_id: anonymousId });CODE
HIGHpackages/cli/src/index.ts813 print({ telemetry: "off", anonymous_id: "none", discarded_anonymous_id: prior.anonymousId, delete_sent_data: TELEMETCODE
HIGHpackages/cli/src/index.ts816 print({ telemetry: "off", anonymous_id: "none" });CODE
HIGHpackages/cli/src/index.ts2006 print({ skills: AGENT_SKILL_METADATA, suites: AGENT_SKILL_SUITES });CODE
HIGHpackages/cli/src/index.ts2713 print({CODE
HIGHpackages/cli/src/index.ts3363 print({CODE
HIGHpackages/cli/src/index.ts8342 return FalseCODE
HIGHpackages/cli/src/index.ts8345 return FalseCODE
HIGHpackages/cli/src/index.ts9695 print({ ...result, repair: result.components.shared_runtime ? "caveman start" : "caveman setup --install", trust: "nCODE
HIGHpackages/cli/src/index.ts9717 print({CODE
HIGHpackages/cli/src/index.ts10796 print({ authenticated: true, baseURL, organization_id: organizationId ?? null, project_id: credentials.project_iCODE
HIGHpackages/cli/src/index.ts10808 print({ authenticated: true, baseURL, gateway_url: gateway || null, organization_id: organizationId ?? null, tokenCODE
HIGHpackages/cli/src/index.ts15277 print(receipt);CODE
HIGHpackages/cli/src/index.ts15341 print(explanation);CODE
HIGHpackages/cli/src/index.ts18237 print({ linked: "codex", source: "local_rate_limits", token_store: "none", refresh });CODE
HIGHpackages/cli/src/index.ts18265 print({ linked: "claude", token_store: tokenStore, basis: "linked_api", refresh });CODE
HIGHpackages/cli/src/index.ts18897 print({ ...view, native_integrations: integrations });CODE
HIGHpackages/cli/src/index.ts18920 print({CODE
HIGHpackages/cli/src/index.ts18946 print(body);CODE
HIGHpackages/cli/src/index.ts18978 print(await response.json());CODE
HIGHpackages/cli/src/index.ts19025 print(body);CODE
HIGHpackages/cli/src/index.ts19330 print({CODE
HIGHpackages/cli/src/index.ts19369 print({ exported: out, receipts: Array.isArray(bundle.receipts) ? bundle.receipts.length : 0 });CODE
Excessive Try-Catch Wrapping115 hits · 131 pts
SeverityFileLineSnippetContext
LOWevals/measure.py42 except Exception:CODE
LOWevals/plot.py29 except Exception:CODE
LOWevals/score_fidelity.py32 except Exception:CODE
LOWevals/llm_run.py74 except Exception:CODE
LOWevals/llm_run.py170 except Exception:CODE
MEDIUMevals/llm_run.py163def claude_version() -> str:CODE
LOW.blocks/release-smoke.py151 except Exception as error:CODE
LOWplugins/caveman/skills/caveman-compress/scripts/cli.py20 except Exception:CODE
LOWplugins/caveman/skills/caveman-compress/scripts/cli.py80 except Exception as e:CODE
LOW…ns/caveman/skills/caveman-compress/scripts/compress.py39 except Exception:CODE
LOW…ns/caveman/skills/caveman-compress/scripts/compress.py294 except Exception:CODE
LOW…ns/caveman/skills/caveman-compress/scripts/compress.py358 except Exception:CODE
LOWtests/verify_repo.py25 except Exception:CODE
LOW⚡tests/test_mode_tracker.py91 except Exception:CODE
LOWtests/middleware-e2e/drive_py.py95 except Exception: # report the failure as the JSON line the harness readsCODE
LOWbenchmarks/run.py21 except Exception:CODE
LOW…ckages/middleware/python/caveman_middleware/strands.py60 except Exception as error: # Decision 4: the native model still receives the caller's messagesCODE
MEDIUM…ckages/middleware/python/caveman_middleware/strands.py57def _prepare(self, messages, *args, **kwargs):CODE
LOW…es/middleware/python/caveman_middleware/llama_index.py479 except Exception as error: # Decision 4: the provider still receives the caller's messagesCODE
LOW…es/middleware/python/caveman_middleware/llama_index.py485 except Exception as error:CODE
LOW…es/middleware/python/caveman_middleware/llama_index.py540 except Exception:CODE
LOW…es/middleware/python/caveman_middleware/llama_index.py813 except Exception as error: # Decision 4: the synthesizer keeps the original nodesCODE
LOW…es/middleware/python/caveman_middleware/llama_index.py827 except Exception as error:CODE
LOW…es/middleware/python/caveman_middleware/llama_index.py850 except Exception:CODE
MEDIUM…es/middleware/python/caveman_middleware/llama_index.py476def _prepare(self, messages, tools, kwargs):CODE
MEDIUM…es/middleware/python/caveman_middleware/llama_index.py482def _aprepare(self, messages, tools, kwargs):CODE
LOWpackages/middleware/python/caveman_middleware/crewai.py201 except Exception as error: # Decision 4: the delegate still receives the executor's messagesCODE
MEDIUMpackages/middleware/python/caveman_middleware/crewai.py198def _guarded_options(self, messages, *args):CODE
LOWpackages/middleware/python/caveman_middleware/asgi.py123 except Exception as error: # Decision 4: the application still handles the requestCODE
LOW⚡packages/middleware/python/caveman_middleware/agno.py268 except Exception as error: # Decision 4: invalid_scope was already warned; other failures pass throughCODE
LOWpackages/middleware/python/caveman_middleware/agno.py370 except Exception as error:CODE
LOWpackages/middleware/python/caveman_middleware/agno.py382 except Exception as error:CODE
MEDIUMpackages/middleware/python/caveman_middleware/agno.py373def _aproject(self, name, messages, args, kwargs, runtime):CODE
LOWpackages/middleware/python/caveman_middleware/google.py84 except Exception as error: # Decision 4CODE
LOWpackages/middleware/python/caveman_middleware/google.py507 except Exception as error: # Decision 4CODE
LOWpackages/middleware/python/caveman_middleware/google.py546 except Exception as error: # Decision 4CODE
MEDIUMpackages/middleware/python/caveman_middleware/google.py439def __iter__(self):CODE
LOW…ckages/middleware/python/caveman_middleware/autogen.py229 except Exception as error: # Decision 4: the model client still receives the caller's messagesCODE
MEDIUM…ckages/middleware/python/caveman_middleware/autogen.py226def _prepare(self, messages, *args):CODE
LOW…ckages/middleware/python/caveman_middleware/litellm.py192 except Exception as error: # Decision 4CODE
LOW…ckages/middleware/python/caveman_middleware/litellm.py238 except Exception as error: # Decision 4CODE
LOW…ckages/middleware/python/caveman_middleware/litellm.py265 except Exception as error: # LiteLLM would swallow it silently; report it insteadCODE
LOW⚡…ckages/middleware/python/caveman_middleware/litellm.py324 except Exception as error: # accounting never replaces the provider's resultCODE
LOW…ckages/middleware/python/caveman_middleware/litellm.py379 except Exception as error: # Decision 4CODE
MEDIUM…ckages/middleware/python/caveman_middleware/litellm.py313def _finish(self, data, response, event):CODE
LOWpackages/middleware/python/caveman_middleware/openai.py168 except Exception as error: # Decision 4CODE
LOWpackages/middleware/python/caveman_middleware/openai.py195 except Exception as error: # accounting never replaces the provider's resultCODE
MEDIUMpackages/middleware/python/caveman_middleware/openai.py187def finish(result, attempt):CODE
LOW…ckages/middleware/python/caveman_middleware/_native.py299 except Exception:CODE
LOW…ckages/middleware/python/caveman_middleware/_native.py352 except Exception as error: # Decision 4: never break the native callCODE
LOW…ckages/middleware/python/caveman_middleware/_native.py361 except Exception as error:CODE
MEDIUM…ckages/middleware/python/caveman_middleware/_native.py296def registered(self):CODE
MEDIUM…ckages/middleware/python/caveman_middleware/_native.py347def prepare(self, body):CODE
MEDIUM…ckages/middleware/python/caveman_middleware/_native.py356def prepare_async(self, body):CODE
LOWpackages/middleware/python/caveman_middleware/mcp.py102 except Exception as error: # Decision 4: the host keeps the original resultCODE
LOW…ages/middleware/python/caveman_middleware/anthropic.py67 except Exception:CODE
LOW…ages/middleware/python/caveman_middleware/anthropic.py90 except Exception as error: # Decision 4CODE
LOW…ages/middleware/python/caveman_middleware/anthropic.py97 except Exception as error:CODE
MEDIUM…ages/middleware/python/caveman_middleware/anthropic.py85def _passive(self, request):CODE
MEDIUM…ages/middleware/python/caveman_middleware/anthropic.py93def _observe(self, result, attempt):CODE
55 more matches not shown…
Unused Imports118 hits · 106 pts
SeverityFileLineSnippetContext
LOWevals/measure.py25CODE
LOWevals/plot.py12CODE
LOWevals/snapshot_contract.py3CODE
LOWevals/reminder_run.py32CODE
LOWevals/score_fidelity.py20CODE
LOWevals/llm_run.py55CODE
LOWtests/test_eval_fidelity.py4CODE
LOWtests/test_eval_llm_run.py3CODE
LOWtests/verify_repo.py4CODE
LOWtests/verify_repo.py548CODE
LOWtests/verify_repo.py550CODE
LOWtests/test_eval_snapshot_contract.py1CODE
LOWtests/test_eval_measure.py4CODE
LOWshared/provider-catalog/validate_catalog.py2CODE
LOW…ckages/middleware/python/caveman_middleware/strands.py2CODE
LOW…es/middleware/python/caveman_middleware/llama_index.py2CODE
LOW…es/middleware/python/caveman_middleware/llama_index.py12CODE
LOW…ckages/middleware/python/caveman_middleware/_httpx2.py8CODE
LOWpackages/middleware/python/caveman_middleware/crewai.py6CODE
LOWpackages/middleware/python/caveman_middleware/asgi.py7CODE
LOWpackages/middleware/python/caveman_middleware/agno.py2CODE
LOWpackages/middleware/python/caveman_middleware/agno.py12CODE
LOW…kages/middleware/python/caveman_middleware/__init__.py3CODE
LOW…kages/middleware/python/caveman_middleware/__init__.py3CODE
LOW…kages/middleware/python/caveman_middleware/__init__.py3CODE
LOW…kages/middleware/python/caveman_middleware/__init__.py3CODE
LOW…ckages/middleware/python/caveman_middleware/autogen.py7CODE
LOW…ckages/middleware/python/caveman_middleware/autogen.py24CODE
LOW…ckages/middleware/python/caveman_middleware/litellm.py7CODE
LOWpackages/middleware/python/caveman_middleware/openai.py6CODE
LOW…ckages/middleware/python/caveman_middleware/_native.py2CODE
LOWpackages/middleware/python/caveman_middleware/mcp.py11CODE
LOWpackages/middleware/python/caveman_middleware/mcp.py14CODE
LOW…ages/middleware/python/caveman_middleware/anthropic.py6CODE
LOW…ages/middleware/python/caveman_middleware/langchain.py2CODE
LOW…ages/middleware/python/caveman_middleware/langchain.py9CODE
LOW…ages/middleware/python/caveman_middleware/langchain.py18CODE
LOW…es/middleware/python/caveman_middleware/pydantic_ai.py2CODE
LOW…es/middleware/python/caveman_middleware/pydantic_ai.py11CODE
LOWpackages/middleware/python/tests/test_fail_open.py7CODE
LOWpackages/sdk/python/tests/test_trace_continuity.py3CODE
LOWpackages/sdk/python/tests/test_context_pack.py1CODE
LOWpackages/sdk/python/tests/test_exporter.py1CODE
LOWpackages/sdk/python/tests/test_middleware_reports.py5CODE
LOWpackages/sdk/python/tests/test_sdk.py1CODE
LOWpackages/sdk/python/tests/test_shared_context.py3CODE
LOWpackages/sdk/python/tests/test_runtime_policy.py16CODE
LOWpackages/sdk/python/tests/test_runtime_policy.py23CODE
LOWpackages/sdk/python/tests/test_parity.py12CODE
LOWpackages/sdk/python/tests/test_tool_events.py3CODE
LOWpackages/sdk/python/tests/test_task_profile.py1CODE
LOWpackages/sdk/python/tests/test_assembly.py1CODE
LOWpackages/sdk/python/tests/test_middleware_transport.py2CODE
LOWpackages/sdk/python/tests/test_structural.py6CODE
LOWpackages/sdk/python/caveman_cloud/__init__.py1CODE
LOWpackages/sdk/python/caveman_cloud/__init__.py1CODE
LOWpackages/sdk/python/caveman_cloud/__init__.py1CODE
LOWpackages/sdk/python/caveman_cloud/__init__.py1CODE
LOWpackages/sdk/python/caveman_cloud/__init__.py1CODE
LOWpackages/sdk/python/caveman_cloud/__init__.py1CODE
58 more matches not shown…
Deep Nesting57 hits · 48 pts
SeverityFileLineSnippetContext
LOWevals/snapshot_contract.py93CODE
LOWevals/llm_run.py174CODE
LOW.blocks/release-smoke.py33CODE
LOW…ns/caveman/skills/caveman-compress/scripts/validate.py121CODE
LOW…ns/caveman/skills/caveman-compress/scripts/compress.py153CODE
LOW…ns/caveman/skills/caveman-compress/scripts/compress.py801CODE
LOW…gins/caveman/skills/caveman-compress/scripts/detect.py61CODE
LOWtests/verify_repo.py87CODE
LOWtests/verify_repo.py557CODE
LOWtests/test_compress_concurrency.py64CODE
LOWtests/test_compress_concurrency.py111CODE
LOWtests/test_compress_concurrency.py131CODE
LOWtests/test_compress_concurrency.py147CODE
LOWtests/test_compress_concurrency.py162CODE
LOWtests/test_compress_concurrency.py188CODE
LOWtests/test_compress_concurrency.py200CODE
LOWtests/test_compress_concurrency.py212CODE
LOW…ckages/middleware/python/caveman_middleware/strands.py65CODE
LOW…es/middleware/python/caveman_middleware/llama_index.py242CODE
LOW…es/middleware/python/caveman_middleware/llama_index.py579CODE
LOW…es/middleware/python/caveman_middleware/llama_index.py614CODE
LOWpackages/middleware/python/caveman_middleware/agno.py442CODE
LOWpackages/middleware/python/caveman_middleware/agno.py476CODE
LOWpackages/middleware/python/caveman_middleware/google.py282CODE
LOWpackages/middleware/python/caveman_middleware/google.py356CODE
LOWpackages/middleware/python/caveman_middleware/google.py398CODE
LOW…ckages/middleware/python/caveman_middleware/autogen.py330CODE
LOW…ckages/middleware/python/caveman_middleware/litellm.py268CODE
LOWpackages/middleware/python/caveman_middleware/_usage.py46CODE
LOW…ckages/middleware/python/caveman_middleware/_native.py39CODE
LOW…ckages/middleware/python/caveman_middleware/_native.py107CODE
LOW…ckages/middleware/python/caveman_middleware/_native.py51CODE
LOW…es/middleware/python/caveman_middleware/pydantic_ai.py73CODE
LOWpackages/middleware/python/tests/test_contracts.py162CODE
LOWpackages/middleware/python/tests/test_openai_native.py106CODE
LOWpackages/sdk/python/tests/test_middleware_preflight.py62CODE
LOWpackages/sdk/python/tests/test_exporter.py20CODE
LOWpackages/sdk/python/tests/test_middleware_v1_1.py147CODE
LOWpackages/sdk/python/tests/test_middleware_v1_1.py161CODE
LOWpackages/sdk/python/tests/test_middleware_v1_1.py178CODE
LOWpackages/sdk/python/tests/test_middleware_v1_1.py214CODE
LOWpackages/sdk/python/tests/test_middleware_v1_1.py239CODE
LOWpackages/sdk/python/tests/test_middleware_v1_1.py261CODE
LOWpackages/sdk/python/tests/test_middleware_v1_1.py285CODE
LOWpackages/sdk/python/tests/test_middleware_transport.py245CODE
LOWpackages/sdk/python/tests/test_middleware_transport.py254CODE
LOWpackages/sdk/python/caveman_cloud/core.py425CODE
LOWpackages/sdk/python/caveman_cloud/core.py1778CODE
LOWpackages/sdk/python/caveman_cloud/core.py2340CODE
LOW…kages/sdk/python/caveman_cloud/middleware/transport.py168CODE
LOWpackages/sdk/python/caveman_cloud/middleware/runtime.py435CODE
LOWpackages/sdk/python/caveman_cloud/middleware/runtime.py592CODE
LOWpackages/sdk/python/caveman_cloud/middleware/runtime.py678CODE
LOWskills/caveman-compress/scripts/validate.py121CODE
LOWskills/caveman-compress/scripts/compress.py153CODE
LOWskills/caveman-compress/scripts/compress.py801CODE
LOWskills/caveman-compress/scripts/detect.py61CODE
Hallucination Indicators3 hits · 45 pts
SeverityFileLineSnippetContext
CRITICAL⚡engine/compressors/testreport_test.go26 at org.junit.internal.runners.model.ReflectiveCallable.run(ReflectiveCallable.java:12)CODE
CRITICAL⚡engine/compressors/testreport_test.go27 at org.junit.runners.model.FrameworkMethod.invokeExplosively(FrameworkMethod.java:47)CODE
CRITICAL⚡engine/compressors/testreport_test.go28 at org.junit.internal.runners.statements.InvokeMethod.evaluate(InvokeMethod.java:17)CODE
Synthetic Comment Markers3 hits · 34 pts
SeverityFileLineSnippetContext
HIGH⚡packages/cli/src/index.ts4086// - valid entitlement (now < expires_at): compression as requested, "entitled".COMMENT
HIGH⚡packages/cli/src/index.ts4087// - lapsed ≤7 days: compression as requested, "grace" (caller kicks a refresh).COMMENT
HIGH⚡packages/cli/src/index.ts4088// - no/expired/unreadable entitlement: compression as requested, "unentitled".COMMENT
AI Slop Vocabulary12 hits · 32 pts
SeverityFileLineSnippetContext
MEDIUMtests/middleware-e2e/drive_py.py95 except Exception: # report the failure as the JSON line the harness readsCODE
MEDIUMbenchmarks/render_charts.py57 # The current harness has three arms. Plot the skill's contributionCOMMENT
MEDIUMbenchmarks/run.py61# the eval harness uses (evals/llm_run.py TERSE_PREFIX) so the two agree.COMMENT
MEDIUMpackages/cli/src/agent-skills.generated.ts20 "megacave": "---\nname: megacave\ndescription: >\n Caveman in Classical Chinese: 文言文 register, far fewer characters, CODE
MEDIUMpackages/cli/src/index.ts14718// CLI, robust to PATH: a resolved `caveman`/`cave`, else this very script's node.COMMENT
MEDIUM.github/workflows/engine-ci.yml236 # and is excluded until its harness is ported, not because coverage isCOMMENT
MEDIUMengine/compressors/json.go55 anomalyZ float64 // robust |z| (MAD-based) at/above which an item is an anomalyCODE
MEDIUMengine/compressors/json.go175 // Numeric anomalies (robust MAD-z per numeric leaf path + element length).COMMENT
MEDIUMengine/compressors/json.go330// computes a robust MAD-based z over the elements that have that path and flagsCOMMENT
MEDIUMengine/compressors/json.go410// medianAndScale returns the median and a robust scale (MAD) of xs, falling backCOMMENT
MEDIUMengine/compressors/json.go597// and accepts it when the mean shift across k exceeds zThresh robust scales.COMMENT
MEDIUMsrc/hooks/caveman-mode-tracker.js434// Same failure, output side (#397): the harness can close its end of ourCOMMENT
Modern AI Meta-Vocabulary8 hits · 23 pts
SeverityFileLineSnippetContext
MEDIUM⚡evals/snapshots/results.json32 "The main question is how many consumers should get each message.\n\n## Queue (point-to-point)\n**Each message is CODE
MEDIUMproxy/providers/gemini/explicit_cache.go12// runtime scaffold. It is not a current policy flag or optimization label.COMMENT
MEDIUMproxy/providers/anthropic/cogs_bench_test.go7// body, which prices a million agentic requests at roughly EUR 0.13 of ScalewayCOMMENT
MEDIUMproxy/providers/anthropic/cogs_bench_test.go43// message history — the shape an agentic caller actually sends.COMMENT
MEDIUMproxy/internal/gateway/proxy.go1743// The historical gemini-explicit-cache scaffold is permanently absent: it neverCOMMENT
MEDIUMshared/platform/optimizers/optimizers_test.go86 // gemini-explicit-cache was a runtime-policy scaffold, never a money-bearingCOMMENT
MEDIUMshared/platform/optimizers/optimizers.go27// telemetry join key. The old scaffold never created a cachedContents resource,COMMENT
MEDIUMpackages/cli/src/agent-skills.generated.ts17 "caveman-manage": "---\nname: caveman-manage\ndescription: >\n Inspect Caveman Cloud's experiment lifecycle and blockCODE
Cross-Language Confusion2 hits · 14 pts
SeverityFileLineSnippetContext
HIGHevals/llm_run.py51 (unfinished cells null), leaves results.json alone andSTRING
HIGH⚡tests/test_eval_snapshot_contract.py127 # Proves a failed (null) or negative usage cell is not evidence.COMMENT
Redundant / Tautological Comments5 hits · 8 pts
SeverityFileLineSnippetContext
LOWplugins/caveman/skills/caveman-compress/scripts/cli.py56 # Check if compressibleCOMMENT
LOW…ns/caveman/skills/caveman-compress/scripts/compress.py931 # Check if backup already exists to prevent accidental overwritingCOMMENT
LOWskills/caveman-compress/scripts/cli.py56 # Check if compressibleCOMMENT
LOWskills/caveman-compress/scripts/compress.py931 # Check if backup already exists to prevent accidental overwritingCOMMENT
LOWsrc/hooks/install.sh77# Check if already installed (unless --force). Older installs only had two hookCOMMENT
Fake / Example Data6 hits · 7 pts
SeverityFileLineSnippetContext
LOWproxy/internal/identity/identity_test.go142 "mtls unknown kind": mapFile(entry("mtls:email:a@example.org", `["*"]`)),CODE
LOWtests/test_hooks.py330 home = Path(tmp) / "Jane Doe"CODE
LOWpackages/middleware/python/tests/test_behavior.py546@pytest.mark.parametrize("session", ["user@example.com", "user 42 / chat #7"])CODE
LOW⚡engine/compressors/testdata/axtree_cdp_local_page.json614 "type": "placeholder",CODE
LOW⚡engine/compressors/testdata/axtree_cdp_local_page.json615 "attribute": "placeholder",CODE
LOW⚡engine/compressors/testdata/axtree_cdp_local_page.json619 "type": "placeholder",CODE
Verbosity Indicators4 hits · 6 pts
SeverityFileLineSnippetContext
LOW…ns/caveman/skills/caveman-compress/scripts/compress.py949 # Step 1: Compress (body only, frontmatter excluded)COMMENT
LOW…ns/caveman/skills/caveman-compress/scripts/compress.py1005 # Step 2: Validate + Retry. Each candidate is staged and validated nextCOMMENT
LOWskills/caveman-compress/scripts/compress.py949 # Step 1: Compress (body only, frontmatter excluded)COMMENT
LOWskills/caveman-compress/scripts/compress.py1005 # Step 2: Validate + Retry. Each candidate is staged and validated nextCOMMENT
Structural Annotation Overuse4 hits · 6 pts
SeverityFileLineSnippetContext
LOW…ns/caveman/skills/caveman-compress/scripts/compress.py949 # Step 1: Compress (body only, frontmatter excluded)COMMENT
LOW…ns/caveman/skills/caveman-compress/scripts/compress.py1005 # Step 2: Validate + Retry. Each candidate is staged and validated nextCOMMENT
LOWskills/caveman-compress/scripts/compress.py949 # Step 1: Compress (body only, frontmatter excluded)COMMENT
LOWskills/caveman-compress/scripts/compress.py1005 # Step 2: Validate + Retry. Each candidate is staged and validated nextCOMMENT
Modern Structural Boilerplate5 hits · 5 pts
SeverityFileLineSnippetContext
LOW…ns/caveman/skills/caveman-compress/scripts/__init__.py7__all__ = ["cli", "compress", "detect", "validate"]CODE
LOW…kages/middleware/python/caveman_middleware/__init__.py5__all__ = ["COMPATIBILITY", "preflight", "ready"]CODE
LOWpackages/sdk/python/caveman_cloud/__init__.py28__all__ = [CODE
LOW…ckages/sdk/python/caveman_cloud/middleware/__init__.py18__all__ = [CODE
LOWskills/caveman-compress/scripts/__init__.py7__all__ = ["cli", "compress", "detect", "validate"]CODE
Slop Phrases1 hit · 3 pts
SeverityFileLineSnippetContext
MEDIUM⚡evals/snapshots/results.json32 "The main question is how many consumers should get each message.\n\n## Queue (point-to-point)\n**Each message is CODE
Overly Generic Function Names3 hits · 3 pts
SeverityFileLineSnippetContext
LOW…ckages/middleware/python/caveman_middleware/_httpx2.py58 def handle_request(self, request):CODE
LOWpackages/middleware/python/caveman_middleware/google.py497 def handle_request(self, request):CODE
LOWpackages/sdk/python/tests/test_transport.py29 def handle_request(self):CODE
AI Structural Patterns3 hits · 2 pts
SeverityFileLineSnippetContext
LOWpackages/sdk/python/caveman_cloud/core.py1661CODE
LOWpackages/sdk/python/caveman_cloud/middleware/runtime.py206CODE
LOWpackages/sdk/python/caveman_cloud/middleware/runtime.py435CODE