SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillsscan-docs-index

scan-docs-index

Writing & research

View repository

Updated 10-15 21:35

About this skill

Codex compatibility note:

  • Invoke repository skills with $skill-name in Codex; this mirrored copy rewrites legacy Claude /skill-name references.
  • Task tracker mandate: BEFORE executing any workflow or skill step, create/update task tracking for all steps and keep it synchronized as progress changes.
  • User-question prompts mean to ask the user directly in Codex.
  • Ignore Claude-specific mode-switch instructions when they appear.
  • Strict execution contract: when a user explicitly invokes a skill, execute that skill protocol as written.
  • Subagent authorization: when a skill is user-invoked or AI-detected and its protocol requires subagents, that skill activation authorizes use of the required spawn_agent subagent(s) for that task.
  • Do not skip, reorder, or merge protocol steps unless the user explicitly approves the deviation first.
  • For workflow skills, execute each listed child-skill step explicitly and report step-by-step evidence.
  • If a required step/tool cannot run in this environment, stop and ask the user before adapting.

Codex Project-Reference Loading (No Hooks)

Codex does not receive Claude hook-based doc injection. When coding, planning, debugging, testing, or reviewing, open project docs explicitly using this routing.

Always read:

  • docs/project-config.json (project-specific paths, commands, modules, and workflow/test settings)
  • docs/project-reference/docs-index-reference.md (routes to the full docs/project-reference/* catalog)
  • docs/project-reference/lessons.md (always-on guardrails and anti-patterns)

Situation-based docs:

  • Backend/CQRS/API/domain/entity changes: backend-patterns-reference.md, domain-entities-reference.md, project-structure-reference.md
  • Frontend/UI/styling/design-system: frontend-patterns-reference.md, scss-styling-guide.md, design-system/README.md
  • Spec/test-case planning or TC mapping: feature-docs-reference.md
  • Integration test implementation/review: integration-test-reference.md
  • E2E test implementation/review: e2e-test-reference.md
  • Code review/audit work: code-review-rules.md plus domain docs above based on changed files

Do not read all docs blindly. Start from docs-index-reference.md, then open only relevant files for the task.

Quick Summary

Goal: Scan the project's docs/ directory → populate docs/project-reference/docs-index-reference.md with accurate documentation tree, file counts by category, doc relationships, and keyword-to-doc lookup table.

Workflow:

  1. Classify — Detect doc organization type, scan mode
  2. Scan — Count docs by category, trace relationships, build lookup
  3. Generate — Build/update reference doc with verified counts and paths
  4. Fresh-Eyes — Round 2 verification validates all counts and paths

Key Rules:

  • Generic — discover everything dynamically, never hardcode project-specific values
  • ALL file counts must be verified via glob, not copied from existing content MUST ATTENTION evidence gate required for EVERY count claim — never estimate
Scan Docs Index

Phase 0: Classify Doc Organization

Before any other step, run in parallel:

  1. Read docs/project-reference/docs-index-reference.md

    • Detect mode: init (placeholder only) or sync (has real content)
    • In sync: note which sections exist and current file counts to diff

  2. Detect documentation organization type:

Signal Type Scan Approach
Structured docs/{category}/ directories Structured hierarchy Scan per-category with phase table below
Single flat docs/ with all files Flat structure Single glob, categorize by filename prefix
wiki/ or external doc system Wiki-based Scan wiki directory, note external docs
Mix of docs + inline README.md files Hybrid Scan both docs/ and source-embedded READMEs
  1. Load service paths from docs/project-config.json if available

Evidence gate: Confidence <60% on organization type → ask user, DO NOT guess structure.

Phase 1: Plan

Create task tracking entries for each scan dimension. Do not start Phase 2 without tasks created.

Phase 2: Scan Documentation Tree

Write findings incrementally after each category — NEVER batch at end.

Think (Coverage dimension): Which directories exist under docs/? Which ones have content vs are empty/stub?

Think (Accuracy dimension): For each count in the existing doc, does the actual glob match? What's the delta?

Think (Completeness dimension): Are there markdown files outside documented directories (e.g., in src/, .claude/, project root)? Are those included in any category?

Think (Discovery dimension): Which files don't fit any existing category? Where do they go?

Root-Level Docs

  • Glob for *.md in project root (README.md, CLAUDE.md, CHANGELOG.md, etc.)
  • Record each with one-line purpose description
  • Evidence gate: File count verified via glob — NEVER estimate

docs/ Directory

Scan each subdirectory with verified glob counts:

Category Glob Pattern What to Extract
project-reference/ docs/project-reference/**/*.md File count (verified), list with purposes
business-features/ docs/business-features/**/*.md Count per app, feature count
operations docs/getting-started.md, docs/deployment.md, etc. File count, list
design-system/ docs/design-system/**/*.md or docs/project-reference/design-system/**/*.md File count, app mapping
specs/ docs/specs/**/*.md File count, module coverage
architecture-decisions/ docs/architecture-decisions/**/*.md ADR count
templates/ docs/templates/**/*.md Template count and types
release-notes/ docs/release-notes/**/*.md File count

Uncategorized files discovery rule: After scanning all categories above, run a broad glob for docs/**/*.md and diff against the union of all category globs. Files in the diff are uncategorized — create a separate "Uncategorized / Other" section for them. NEVER silently omit files.

.claude/docs/

  • Glob for .claude/docs/**/*.md — count and categorize
  • Glob for .claude/skills/**/*.md — count skills

Phase 3: Build Doc Relationship Map

Think: Which docs serve as entry points (README → guide chains)? Which docs are referenced from multiple places? Which are isolated?

Trace key doc relationships by grepping for markdown links between docs:

  • Entry points (README → getting-started → deployment chain)
  • CLAUDE.md → reference doc pointers
  • Which docs link to which (cross-references)

Phase 4: Build Lookup Table

For each docs/business-features/{App}/ directory:

  • Extract the app name and key business domain keywords
  • Map keywords → directory path for the lookup table

For each docs/project-reference/*.md:

  • Extract the domain covered
  • Map keywords → file path

Phase 5: Fresh-Eyes Verification

Spawn a fresh sub-agent (zero memory) to independently verify:

  1. Sample 5 file paths from each category — do they exist? (Glob check)
  2. Does the total count for each category match a fresh glob of that pattern?
  3. Are there any files in docs/**/*.md that appear in no category? (Run the diff)
  4. Does the lookup table have entries for all documented categories?
  5. Are there duplicate entries in the lookup table (same path, different keyword)?
  6. Are uncategorized files documented in a separate section?

Proceed to Phase 6 only after fresh-eyes verification passes — do NOT advance while it is still failing.

Phase 6: Generate Reference Doc

Write to docs/project-reference/docs-index-reference.md with sections:

<!-- Last scanned: {YYYY-MM-DD} -->

# Documentation Index Reference

> Auto-generated by `$scan-docs-index`. Do not edit manually.

## Documentation System

{total} markdown files across {N} categories. Last scanned: {date}.

## Documentation Graph

{ASCII tree with counts — counts from verified globs only}

## Key Doc Relationships

{ASCII relationship diagram — entry points and cross-references}

## Doc Lookup Guide

{keyword → path table}

## Uncategorized Files

{Files found by broad glob not in any category — with paths}

[IMPORTANT] Use task tracking to break ALL work into small tasks BEFORE starting.

Prerequisites: MUST ATTENTION READ before executing:

Critical Thinking Mindset — Every claim needs traced proof, confidence >80% to act. Anti-hallucination: Never present guess as fact — cite sources, admit uncertainty, self-check output, cross-reference independently. Certainty without evidence = root of all hallucination.

Scan & Update Reference Doc — Surgical updates only, NEVER full rewrite.

  1. Read existing doc first — understand structure and manual annotations
  2. Detect mode: Placeholder (headings only) → Init. Has content → Sync.
  3. Scan codebase (grep/glob) for current state
  4. Diff findings vs doc — identify stale sections only
  5. Update ONLY diverged sections. Preserve manual annotations.
  6. Update metadata (date, version) in frontmatter/header
  7. NEVER rewrite entire doc. NEVER remove sections without evidence obsolete.

Output Quality — Token efficiency without sacrificing quality.

  1. No inventories/counts — stale instantly
  2. No directory trees — use 1-line path conventions
  3. No TOCs — AI reads linearly
  4. One example per pattern — only if non-obvious
  5. Lead with answer, not reasoning
  6. Sacrifice grammar for concision in reports
  7. Unresolved questions at end

AI Mistake Prevention — Failure modes to avoid:

Verify AI-generated content against actual code. AI hallucinates file paths and counts. Glob to confirm existence before documenting. Trace full dependency chain after edits. Always trace full chain. Surface ambiguity before coding. NEVER pick silently. Update docs that embed canonical data when source changes. Docs inlining counts go stale silently.

IMPORTANT MUST ATTENTION read existing doc first, scan codebase, diff, surgical update only. Never rewrite entire doc.

IMPORTANT MUST ATTENTION output quality: no counts/trees/TOCs in the skill output itself, 1 example per pattern, lead with answer.

MUST ATTENTION apply critical thinking — every claim needs traced proof, confidence >80% to act. Anti-hallucination: never present guess as fact.

MUST ATTENTION apply AI mistake prevention — holistic-first debugging, fix at responsible layer, surface ambiguity before coding, re-read files after compaction.

Closing Reminders

IMPORTANT MUST ATTENTION break work into small task tracking tasks BEFORE starting IMPORTANT MUST ATTENTION detect doc organization type in Phase 0 — scan approach depends on it IMPORTANT MUST ATTENTION evidence gate for EVERY count — glob to verify, NEVER estimate or copy from existing content IMPORTANT MUST ATTENTION write findings incrementally after each category — NEVER batch at end IMPORTANT MUST ATTENTION run uncategorized file discovery — NEVER silently omit files that don't fit categories IMPORTANT MUST ATTENTION Phase 5 fresh-eyes verification is mandatory before writing final doc

Anti-Rationalization:

Evasion Rebuttal
"Count looks right from existing doc, skip glob" EVERY count requires fresh glob verification — no exceptions
"Only need to check 3 paths" Phase 5 has 6 specific checks — sample across all categories
"All files fit into existing categories" Run the uncategorized discovery diff — NEVER assume full coverage
"Skip Round 2 even when Round 1 found issues" Clean Round 1 ends the scan. When issues exist, fresh-eyes mandatory after fixing — main agent's counts carry confirmation bias.
"Lookup table doesn't need all keywords" Map keywords for EVERY documented category, not just top-level

[TASK-PLANNING] Before acting, analyze task scope and break into small todo tasks and sub-tasks using task tracking.

Hookless Prompt Protocol Mirror (Auto-Synced)

Source: .claude/hooks/lib/prompt-injections.cjs + .claude/.ck.json

[WORKFLOW-EXECUTION-PROTOCOL] [BLOCKING] Workflow Execution Protocol — MANDATORY IMPORTANT MUST CRITICAL. Do not skip for any reason.

Generic portability boundary: Reusable skills and protocol text stay project-neutral; project-specific conventions are discovered from docs/project-config.json and docs/project-reference/. Apply shared AI-SDD from shared/sdd-artifact-contract.md. Read docs/project-config.json and docs/project-reference/docs-index-reference.md, then open the project reference docs named there. Any supported AI tool may execute when this shared context and local docs are available.

  1. DETECT: Match prompt against workflow catalog
  2. ANALYZE: Find best-match workflow AND evaluate if a custom step combination would fit better
  3. ASK (REQUIRED FORMAT): Use a direct user question with this structure unless the user explicitly invoked a workflow/skill and the local protocol treats explicit invocation as confirmation:
    • Question: "Which workflow do you want to activate?"
    • Option 1: "Activate [BestMatch Workflow] (Recommended)"
    • Option 2: "Activate custom workflow: [step1 → step2 → ...]" (include one-line rationale)
  4. ACTIVATE (if confirmed): Call $workflow-start <workflowId> for standard; sequence custom steps manually
  5. CREATE TASKS: task tracking for ALL workflow steps
  6. EXECUTE: Follow each step in sequence [CRITICAL-THINKING-MINDSET] Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination principle: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination. AI Attention principle (Primacy-Recency): Put the 3 most critical rules at both top and bottom of long prompts/protocols so instruction adherence survives long context windows. Goal-driven execution: Define success criteria first, loop until verified, and stop only when observable checks pass. Tests verify intent: Tests must protect business rules/invariants and fail when the protected intent breaks, not only mirror current behavior.

[LESSON-LEARNED-REMINDER] [BLOCKING] Task Planning & Continuous Improvement — MANDATORY. Do not skip.

Break work into small tasks (task tracking) before starting. Add final task: "Analyze AI mistakes & lessons learned".

Extract lessons — ROOT CAUSE ONLY, not symptom fixes:

  1. Name the FAILURE MODE (reasoning/assumption failure), not symptom — "assumed API existed without reading source" not "used wrong enum value".
  2. Generality test: does this failure mode apply to ≥3 contexts/codebases? If not, abstract one level up.
  3. Write as a universal rule — strip project-specific names/paths/classes. Useful on any codebase.
  4. Consolidate: multiple mistakes sharing one failure mode → ONE lesson.
  5. Recurrence gate: "Would this recur in future session WITHOUT this reminder?" — No → skip $learn.
  6. Auto-fix gate: "Could $code-review/$code-simplifier/$security/$lint catch this?" — Yes → improve review skill instead.
  7. BOTH gates pass → ask user to run $learn. [TASK-PLANNING] [MANDATORY] BEFORE executing any workflow or skill step, create/update task tracking for all planned steps, then keep it synchronized as each step starts/completes.