skills/glossary/SKILL.md
Extract ubiquitous language from Slack and generate domain glossary. Use when user mentions 用語集, glossary, ユビキタス言語, ドメイン用語, Slack から用語.
npx skillsauth add thkt/claude-config glossaryInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Extract domain terms from Slack conversations and generate a glossary draft.
| Variable | Description |
| ----------------- | ----------------------------------------- |
| SLACK_TOKEN | User OAuth Token (xoxp-...) |
| SLACK_WORKSPACE | Workspace subdomain (before .slack.com) |
Verify SLACK_TOKEN is set before any API call.
$1 (optional): Slack channel name(s) or search keywords--refs (optional): Reference materials (md paths, Google Docs/Sheets URLs)Ask via AskUserQuestion:
| Question | Options / Input |
| ------------------- | ----------------------------------------------------------- |
| Slack channels | Channel names or IDs (comma-separated) |
| Time range | 1 week / 1 month / 3 months / 6 months |
| Search keywords | Optional domain keywords to focus extraction |
| Reference materials | md paths, Google Docs/Sheets URLs (comma-separated or none) |
| Output path | Default: docs/domain/glossary.md |
| Existing glossary? | Path to existing file (merge mode) or none (new) |
Fetch channel history, keyword search, thread replies via Slack API. Paginate
with next_cursor (max 5 pages per channel).
→ curl examples: reference.md
When reference materials are specified, load them as additional context for term extraction.
| Source | How to load |
| -------------- | ------------------------------------- |
| Local md files | Read tool |
| Google Docs | google-workspace skill (gcloud CLI) |
| Google Sheets | google-workspace skill (gcloud CLI) |
Reference materials are NOT embedded into kiku's DB. They serve as cross-reference context during Phase 3 term extraction — helping the LLM validate discovered terms against formal documentation.
Analyze harvested conversations in batches. For each batch, extract:
| Extract | Signal patterns | | ----------------- | ------------------------------------------------------------------- | | Term candidates | Nouns/phrases repeated across multiple messages | | Definitions | "X means Y", "X is Y", "X = Y", "X って Y のこと" | | Synonyms | "X (or Y)", "X aka Y", "X とも言う" | | Corrections | "X じゃなくて Y", "not X but Y" — shows which term is canonical | | Code references | Backtick-wrapped identifiers that map to business terms | | Ambiguity signals | "X って何?", "what do you mean by X" — terms needing clarification |
For each batch, produce a table with: Term, Definition, Code Mapping, Confidence, Evidence, Ref Mapping, Synonyms. Cross-reference against loaded reference materials when available.
→ Output template: reference.md
| Level | Criteria | | ------ | ---------------------------------------------- | | high | Explicit definition found in conversation | | medium | Inferred from usage context across 2+ messages | | low | Single mention, definition is speculative |
If a project codebase is available, cross-reference extracted terms:
code_mapping field in extraction outputUse Grep for exact identifier matches.
Write glossary to output path. Includes Terms table, Synonyms table, and Ambiguous Terms section. Merge mode: diff against existing glossary before writing.
→ Templates: reference.md
/glossary #dev #product
/glossary #dev --refs docs/design.md,https://docs.google.com/document/d/xxx
/glossary 決済 注文 ユーザー
/glossary
| Field | Standard |
| ---------- | ------------------------------------------------- |
| Definition | Cite Slack source: slack:#{channel}/p{ts} |
| Confidence | Mark per confidence level table |
| Ambiguity | List in separate section, never guess definitions |
| Error | Action |
| ------------------------- | ----------------------------------------------- |
| SLACK_TOKEN not set | Report and stop |
| Channel not accessible | Report channel, suggest checking permissions |
| No terms extracted | Report "No domain terms found in scope" |
| Too many messages (>1000) | Suggest narrowing time range or adding keywords |
| Check | Required | | ------------------------------------ | -------- | | All terms have confidence markers? | Yes | | Ambiguous terms listed separately? | Yes | | Slack sources cited for definitions? | Yes | | User reviewed before final write? | Yes |
tools
Internal helper for /think Step 11. Renders SOW.md + Spec.md as an integrated Astro view and returns a dev server URL.
development
Extract repository spec while detecting bugs, spec gaps, and consistency drift via dual-purpose documentation. OUTCOME.md-axis question-driven exploration with ephemeral output. Do NOT use for code review (use /audit or /polish), feature implementation (use /code), planning only (use /think), or single-bug fix (use /fix).
development
Discover undocumented design decisions and challenge each candidate via critic-design before promotion. Rank by impact and reversibility, produce ADR promotion candidates. Treat each candidate as a position arguing for ADR status, not a fact to be filed. Pairs with audit-adr-drift, which scans existing ADRs for drift against code.
development
Scan ADR Decision sections against current code and report drift with modification direction and priority. Do NOT use for repos without ADRs (use audit-adr-gaps instead).