This report summarizes scanner evidence for HKUSTDial/Supervisor-Skills. SynthScan examined 19,629 lines of code across 138 source files, recording 89 pattern matches distributed across 7 syntactic categories. The legacy adjusted pattern signal is 8.5 (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.
Payload schema: 1.0.0 · Engine: 2.0.0 · Scoring: heuristic-density-v1 · Scan date: 2026-10-31.
Can maintainers explain intent, assumptions, and trade-offs? 84 finding(s).
Are imports declared and package/API names verified? 2 finding(s).
Are wrappers, guards, and repeated structures necessary? 5 finding(s).
Which changes need focused human review? 89 finding(s).
Run additional scans to build a history of the legacy pattern signal. Score changes are not a measure of AI adoption or code quality.
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 89 distinct pattern matches across 7 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⚡ | …/05_Vibe_Research/5.1_vibe-research-and-vibe-coding.md | 245 | ### Step 1: Figure Design | COMMENT |
| LOW⚡ | …/05_Vibe_Research/5.1_vibe-research-and-vibe-coding.md | 252 | ### Step 2: Draft Figure Generation | COMMENT |
| LOW⚡ | …/05_Vibe_Research/5.1_vibe-research-and-vibe-coding.md | 256 | ### Step 3: Redraw as a Vector Figure | COMMENT |
| LOW | skills/vibe-research-workflow/SKILL.md | 72 | ### Step 1: Phase classification | COMMENT |
| LOW | skills/vibe-research-workflow/SKILL.md | 77 | ### Step 2: Behavioural rules recap | COMMENT |
| LOW | skills/vibe-research-workflow/SKILL.md | 104 | ### Step 3: Phase-specific procedure | COMMENT |
| LOW | skills/vibe-research-workflow/SKILL.md | 116 | ### Step 4: Tool selection | COMMENT |
| LOW | skills/vibe-research-workflow/SKILL.md | 130 | ### Step 5: Integrity gate | COMMENT |
| LOW | skills/vibe-research-workflow/SKILL.md | 135 | ### Step 6: Output | COMMENT |
| LOW | skills/benchmark-paper-template/SKILL.md | 90 | ## Step 1: Five-pillar completeness table | COMMENT |
| LOW | skills/benchmark-paper-template/SKILL.md | 100 | ## Step 2: Introduction six-part logic chain | COMMENT |
| LOW | skills/benchmark-paper-template/SKILL.md | 111 | ## Step 3: Section outline for §2 to §7 | COMMENT |
| LOW | skills/benchmark-paper-template/SKILL.md | 115 | ## Step 4: Pre-submission self-check | COMMENT |
| LOW | …ls/benchmark-paper-template/references/gap-analysis.md | 5 | ## Step 1: Survey the Evaluation Landscape | COMMENT |
| LOW | …ls/benchmark-paper-template/references/gap-analysis.md | 23 | ## Step 2: Identify the Blind Spot | COMMENT |
| LOW | …ls/benchmark-paper-template/references/gap-analysis.md | 46 | ## Step 3: Validate the Gap | COMMENT |
| LOW | …ls/benchmark-paper-template/references/gap-analysis.md | 57 | ## Step 4: Articulate the Gap Statement | COMMENT |
| LOW | …ls/benchmark-paper-template/references/gap-analysis.md | 68 | ## Step 5: Derive Research Questions | COMMENT |
| LOW | skills/pre-submission-reviewer/SKILL.md | 53 | ### Step 0: paradigm and venue fit | COMMENT |
| LOW | skills/pre-submission-reviewer/SKILL.md | 97 | ### Step 1: Dimension 1 Macro logic review | COMMENT |
| LOW | skills/pre-submission-reviewer/SKILL.md | 143 | ### Step 2: Dimension 2 Writing details review | COMMENT |
| LOW | skills/pre-submission-reviewer/SKILL.md | 155 | ### Step 3: Dimension 3 English grammar review | COMMENT |
| LOW | skills/pre-submission-reviewer/SKILL.md | 171 | ### Step 4: Dimension 4 LaTeX format review | COMMENT |
| LOW | skills/pre-submission-reviewer/SKILL.md | 187 | ### Step 5: Dimension 5 Figure quality review | COMMENT |
| LOW | skills/pre-submission-reviewer/SKILL.md | 202 | ### Step 6: Banned-vocabulary and em-dash scan | COMMENT |
| LOW⚡ | skills/pre-submission-reviewer/SKILL.md | 221 | ### Step 7: Section-by-section review | COMMENT |
| LOW⚡ | skills/pre-submission-reviewer/SKILL.md | 231 | ### Step 8: Integrity gate | COMMENT |
| LOW⚡ | skills/pre-submission-reviewer/SKILL.md | 235 | ### Step 9: Output | COMMENT |
| LOW | skills/deep-research/SKILL.md | 53 | ## Step 0: freeze the research brief | COMMENT |
| LOW | skills/deep-research/references/synthesis-framework.md | 13 | ### Step 1: extraction | COMMENT |
| LOW | skills/deep-research/references/synthesis-framework.md | 24 | ### Step 2: thematic clustering and MECE taxonomy | COMMENT |
| LOW | skills/deep-research/references/synthesis-framework.md | 42 | ### Step 3: serial section synthesis | COMMENT |
| LOW | …ills/paper-writer/references/verification-checklist.md | 13 | ## Step 1: source recall check | COMMENT |
| LOW | …ills/paper-writer/references/verification-checklist.md | 26 | ## Step 2: inline checks | COMMENT |
| LOW | skills/rebuttal-guidance/SKILL.md | 46 | ### Step 1: Inventory concerns | COMMENT |
| LOW | skills/rebuttal-guidance/SKILL.md | 58 | ### Step 2: Classify each concern | COMMENT |
| LOW | skills/rebuttal-guidance/SKILL.md | 66 | ### Step 3: Match reviewer mindset (X) | COMMENT |
| LOW | skills/rebuttal-guidance/SKILL.md | 77 | ### Step 4: Strategy priorities for this mindset (Y) | COMMENT |
| LOW | skills/rebuttal-guidance/SKILL.md | 89 | ### Step 5: Per-concern guidance | COMMENT |
| LOW | skills/rebuttal-guidance/SKILL.md | 108 | ### Step 6: Integrity gate | COMMENT |
| LOW | skills/rebuttal-guidance/SKILL.md | 118 | ### Step 7: Output format | COMMENT |
| LOW | skills/intro-drafter/SKILL.md | 79 | ### Step 1: paper-type positioning (internal) | COMMENT |
| LOW | skills/intro-drafter/SKILL.md | 90 | ### Step 2: plan the six paragraphs (internal) | COMMENT |
| LOW | skills/intro-drafter/SKILL.md | 106 | ### Step 3: the running example (internal) | COMMENT |
| LOW | skills/intro-drafter/SKILL.md | 117 | ### Step 4: alignment checks (internal, before writing) | COMMENT |
| LOW | skills/intro-drafter/SKILL.md | 131 | ### Step 5: write the prose | COMMENT |
| LOW | skills/intro-drafter/SKILL.md | 163 | ### Step 6: a short closing note (optional, three lines at most) | COMMENT |
| LOW | skills/tech-paper-template/SKILL.md | 57 | ### Step 1: Paper-type positioning | COMMENT |
| LOW | skills/tech-paper-template/SKILL.md | 70 | ### Step 2: Fill the thinking template | COMMENT |
| LOW | skills/tech-paper-template/SKILL.md | 87 | ### Step 3: Run four self-consistency checks | COMMENT |
| LOW⚡ | skills/tech-paper-template/SKILL.md | 109 | ### Step 4: Generate methodology outline | COMMENT |
| LOW⚡ | skills/tech-paper-template/SKILL.md | 118 | ### Step 5: Integrity gate | COMMENT |
| LOW⚡ | skills/tech-paper-template/SKILL.md | 123 | ### Step 6: Output | COMMENT |
| LOW | skills/idea-evaluator/SKILL.md | 58 | ### Step 1: First impression and paper-type positioning | COMMENT |
| LOW | skills/idea-evaluator/SKILL.md | 80 | ### Step 2: Fatal-flaws audit (early gate) | COMMENT |
| LOW | skills/idea-evaluator/SKILL.md | 126 | ### Step 3: Lifecycle and capability matching | COMMENT |
| LOW | skills/idea-evaluator/SKILL.md | 139 | ### Step 4: Five-dimension scoring | COMMENT |
| LOW | skills/idea-evaluator/SKILL.md | 173 | ### Step 5: Paradigm-shift probe | COMMENT |
| LOW | skills/idea-evaluator/SKILL.md | 194 | ### Step 6: Feasibility check | COMMENT |
| LOW | skills/idea-evaluator/SKILL.md | 209 | ### Step 7: Integrity gate | COMMENT |
| 9 more matches not shown… | ||||
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| HIGH | …/05_Vibe_Research/5.1_vibe-research-and-vibe-coding.md | 60 | In the latest batch of startups, 25% of the companies reported that 95% of their code was generated by AI. | CODE |
| HIGH⚡ | …/05_Vibe_Research/5.1_vibe-research-and-vibe-coding.md | 258 | Draft figures generated by Gemini / Nano Banana Pro are raster images (PNG) and cannot be directly used in the Paper. | CODE |
| HIGH | …/05_Vibe_Research/5.1_vibe-research-and-vibe-coding.md | 285 | - **Do not directly copy Paper content generated by AI / LLMs**. Students must verify the accuracy of content item by it | CODE |
| HIGH | …ook-en/05_Vibe_Research/5.2_liboyan-practical-notes.md | 37 | - **Current Scale**: The CEO of a world-famous startup incubator reported that in 25% of startups, 95% of the code is ge | CODE |
| HIGH | …ook-en/05_Vibe_Research/5.2_liboyan-practical-notes.md | 98 | Use drawing tools to create a vector figure based on the draft (raster images generated by Gemini or Nano Banana cannot | CODE |
| HIGH | …ook-en/05_Vibe_Research/5.2_liboyan-practical-notes.md | 142 | | 20:57 | Importance of AI-assisted coding for research | 95% of code in 25% of global startups is generated by AI; Vibe | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| MEDIUM | skills/paper-polish/references/ai-tone-guardrails.md | 4 | # AI-tone guardrails and inflated-claim wordlist | COMMENT |
| MEDIUM | skills/paper-writer/references/ai-tone-guardrails.md | 4 | # AI-tone guardrails and inflated-claim wordlist | COMMENT |
| MEDIUM | skills/idea-evaluator/references/paradigm-elephant.md | 84 | ## 5. Worked example: multi-agent collaboration for complex tasks | COMMENT |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | skills/drawio-reconstruction/scripts/crop_assist.py | 95 | CODE | |
| LOW | skills/drawio-reconstruction/scripts/check_drawio.py | 37 | CODE | |
| LOW | skills/drawio-reconstruction/scripts/check_drawio.py | 125 | CODE | |
| LOW | skills/drawio-reconstruction/scripts/check_drawio.py | 245 | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | skills/drawio-reconstruction/scripts/check_drawio.py | 257 | except Exception as exc: | CODE |
| MEDIUM | skills/drawio-reconstruction/scripts/check_drawio.py | 8 | def _float_attr(element, name, default=0.0): | CODE |
| LOW | skills/drawio-reconstruction/scripts/batch_verify.py | 81 | except Exception as exc: | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | scripts/check_shared_sync.py | 18 | CODE | |
| LOW | scripts/lint_skills.py | 17 | CODE |
| Severity | File | Line | Snippet | Context |
|---|---|---|---|---|
| LOW | skills/drawio-reconstruction/scripts/check_drawio.py | 37 | def _semantic_containment_failures(cells_by_id): | CODE |
| LOW | skills/drawio-reconstruction/scripts/check_drawio.py | 225 | def _complex_visual_svg_failures(cells_by_id): | CODE |