菜单
Run sentence-level traceability verification on a blog, then generate viewer.html with interactive evidence panel. No LLM needed — pure Python.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Exhaustively profile a dataset and list ALL possible analyses — distributions, correlations, rankings, trends, group comparisons, anomalies. Reads detective.json for context. Runs after the Detective/Scout, before the Editor. Outputs analyst.json with ana_xx IDs and chart-ready data_tables.
Audit a generated Data2Story blog for build correctness across ALL modalities by ACTUALLY RENDERING it in a real headless browser (when available) — catching blank/0-width charts, broken/oversized media, desktop+mobile overflow, and dead interactions that source inspection cannot see — then having vision agents review the screenshots, playtesting every interactive, and walking the flagship capability contract. Plus VIEW every image for a wrong/AI-faked subject. Fixes layout in place; sends content-correctness problems back to the owning role. Use at Stage 6 after the Programmer assembles index.html and before the Critic; re-runs each revision round. Triggers: a built index.html exists, or you need to confirm charts/media/interactions actually render.
Design the MANDATORY cinematic scroll experience (every blog ships one): a full-bleed, scroll-driven background — primarily from the Scout's verified real imagery — plus the global motion choreography, so the narrative unfolds like a film as the reader scrolls. Owns the BACKGROUND layer + page-level motion (not the Designer's per-section visuals, not the Interaction centerpiece — it stages them inside the scroll). Outputs cinematographer.json (cin_xx scenes). There is no off path: a topic with no sourceable real imagery falls back to a generative-atmosphere or data-driven-spine background, never to a bare column.
Name the piece — re-write the masthead (headline + standfirst + kicker), every section title, and every figure/photo/table caption to a research-driven titling standard, killing the AI-tell patterns (the 'flat statement. flat counter-statement.' two-beat above all) a competent default falls into. Reads editor.md/json + analyst.json + the resolved topic_profile; writes copywriter.json — STRINGS ONLY (masthead{headline,standfirst,kicker}, items{edt_xx:{title}, des_xx:{caption}}), each backed by a real ana_*. Names, never edits: it touches no finding, no number, no data-* id, no layout — so the Verify layer is untouched and the Programmer renders the masthead + figcaptions from copywriter.json verbatim. Runs at Stage 3.5, after the Editor, before the Designer.
Review a finished Data2Story blog against the 5 quality rubric dimensions (visual_design, narrative_pacing, data_method_transparency, claim_data_alignment, insight_value), score each 1-7 with on-page evidence, and emit critic.json with pass/fail + targeted, surgical send-back instructions. Verifies every load-bearing claim/asset against its traceability chain before scoring; applies the caveat-survival, honest-accuracy, and third-party-attribution caps. Does NOT rewrite content — scores and sends back. Use at Stage 6.5 after the Auditor and after verify.py has produced verifier.json; re-runs each revision round. Triggers: a built index.html plus verifier.json exist, or you need to judge whether the article is actually good.
Generate text embeddings via OpenRouter using Qwen3-Embedding-8B.
| name | inspector |
| description | Run sentence-level traceability verification on a blog, then generate viewer.html with interactive evidence panel. No LLM needed — pure Python. |
| argument-hint | ["PROJECT_DIR"] |
| allowed-tools | Bash(*), Read, Write |
Your job is traceability verification. Parse the blog HTML, extract every visible sentence, link each back to its evidence in the role JSONs, and generate a self-contained viewer.html that lets readers inspect the evidence chain.
PROJECT_DIR = first argumentSKILL_DIR = the directory containing this SKILL.md (.../skills/data2story/inspector). Replace SKILL_DIR placeholders with the resolved, quoted path before running Bash. Do not hard-code machine-local paths.index.html, analyst.json, detective.json, designer.json, editor.jsonpython3 SKILL_DIR/scripts/verify.py PROJECT_DIR --log-errors
Produces PROJECT_DIR/inspector.json (sentence→evidence mapping). The output shape (format v3: stats, sentences, unused_ids) is in references/inspector_schema.json.
python3 SKILL_DIR/scripts/generate_viewer.py PROJECT_DIR
This reads index.html + inspector.json and produces viewer.html — a self-contained file that works on file:// (no server needed). You run the script; you do not reimplement it. How it works (tag → style → script ordering), its critical constraints (line-number search, lite JSON, ES5, no fetch), and the viewer UI behavior are documented in references/viewer_internals.json.
python3 SKILL_DIR/scripts/verify.py PROJECT_DIR --log-errors
python3 SKILL_DIR/scripts/generate_viewer.py PROJECT_DIR
PROJECT_DIR/inspector.json — full traceability data (with raw_evidence)PROJECT_DIR/viewer.html — self-contained interactive viewer (works on file://)Done when viewer.html opens directly in a browser (no server), shows the blog with a working 🔍 toggle, and every traced sentence has a visible ID linking to its evidence summary.
Running with --log-errors auto-updates known recurring-case metadata in skills/errors/ for patterns the script detects directly from HTML. Use manual logging only when you discover a new pattern the script does not know yet — and only for Critical/High-severity issues that represent patterns, not one-off mistakes. The full logging process, the error-case markdown template, and the common error types worth logging are in references/error_logging.json.