pm-glossary
Skill2 starsChanged 4 months agoMITView on GitHub
Generate and maintain a project glossary - terms, abbreviations, entities, artifacts, and concepts used across the product lifecycle. Runs progressively throughout the project. Can be called at any time to add terms, review existing entries, or export the full glossary.
In this repository51 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- design-thinkingSkill
- jtbd-buildingSkill
- pm-auditSkill
- pm-business-caseSkill
- pm-business-modelSkill
- pm-business-rule-coreSkill
- pm-business-rule-criticalSkill
- pm-business-rule-governanceSkill
- pm-business-rules-librarySkill
- pm-comms-charterSkill
- pm-decision-modelSkill
- pm-diagramsSkill
- pm-discovery-interviewSkill
- pm-discovery-reportSkill
- pm-domain-analysisSkill
- pm-domain-modelSkill
- pm-entity-registrySkill
- pm-feature-cardSkill
- pm-feature-designSkill
- pm-features-listSkill
- pm-feature-viabilitySkill
- pm-hypothesesSkill
- pm-kotlerSkill
- pm-kpisSkill
- pm-lean-canvasSkill
- pm-market-analysisSkill
- pm-meetingSkill
- pm-mvp-scopeSkill
- pm-onboardingSkill
- pm-open-questionsSkill
- pm-personasSkill
- pm-pitch-deckSkill
- pm-prdSkill
- pm-prioritizeSkill
- pm-privacy-requirementsSkill
- pm-problem-validationSkill
- pm-process-flowsSkill
- pm-product-roadmapSkill
- pm-project-charterSkill
- pm-prototypeSkill
- pm-reconcileSkill
- pm-reconcile-statusSkill
- pm-reverse-extractSkill
- pm-root-causeSkill
- pm-scope-briefSkill
- pm-stakeholder-mapSkill
- pm-stress-testSkill
- pm-stripeSkill
- pm-team-rosterSkill
- pm-tech-feasibilitySkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
15 rules. About 0 are common, 7 are distinctive, and 8 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
8
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 32
No workspace: create just the folder this skill writes into, or write to a path the user names, and say where the file went.
- Very specific: line 33
Read pureinn-variables.md / state.json where a value is actually used, not at the top of the run.
- Distinctive: line 43
After completing a phase, to capture new terminology introduced
- Distinctive: line 44
When a term needs to be clarified or defined precisely
- Very specific: line 64
Check for an existing glossary artifact in pureinn-workspace/[project-slug]/artifacts/glossary.md.
- Distinctive: line 72
If glossary exists: show a summary - how many terms, which domains are covered, when it was last updated.
- Distinctive: line 76
Use AskUserQuestion tool with:
- Very specific: line 83
If user provides explicit intent with the command (e.g., /pm-glossary add "Feature Set, Delivery Stripe, Stripe"), skip the question and …
- Distinctive: line 176
Use in Step 0 to identify gaps in existing glossary.
- Distinctive: line 177
Use in Step 2 to verify output quality before finalizing.
- Very specific: line 198
Read pureinn-variables.md key "Glossary" → get DB URL.
- Very specific: line 199
If blank: ask user, save URL to pureinn-variables.md.
- Very specific: line 200
Check state.json notion_ids.glossary for cached ID.
- Very specific: line 224
Update if entry with matching Name already exists (use mcp__claude_ai_Notion__notion-update-page).
- Distinctive: line 226
After push: report counts (created, updated, errors).
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
PM - Project Glossary
Agent mode (--agent)
Supports --agent: runs autonomously in a subagent, drafts the artifact from existing inputs, and returns a short summary + coverage note.
- No flag → interactive (default); if inputs are heavy, offer agent mode.
--agent→ obey. First check inputs are complete. Anything missing: do NOT invent it - mark[ASSUMED - what/why]in the output and summary. Never hallucinate to fill a gap.
Standalone run
Runs with or without a Pureinn workspace - it needs no other skill's artifact, so it works as a single tool from an install that never ran /pureinn.
- No workspace: create just the folder this skill writes into, or write to a path the user names, and say where the file went. Do not scaffold a project, do not invent a
state.json, and do not send the user to/pureinnfirst. - Read
pureinn-variables.md/state.jsonwhere a value is actually used, not at the top of the run. Missing value: continue and name the capability it costs.
What this skill does
Builds and maintains a living glossary of terms used across the product development lifecycle - from discovery through launch. Captures business terms, domain entities, product artifacts, process concepts, rules, and abbreviations in a structured table.
This skill is continuous - it runs throughout the project, not in a single phase. It can be triggered:
- Manually by the user at any time (
/pm-glossary) - After completing a phase, to capture new terminology introduced
- When a term needs to be clarified or defined precisely
The glossary is additive: each run merges new entries with existing ones. Existing definitions are updated only if explicitly requested.
Dependencies
No hard dependencies. Can run at any point in the project.
Best used after:
pm-project-charter- captures initial product and business terminologypm-domain-model- generates the richest set of domain entities and relationshipspm-business-rules-library,pm-feature-design- introduces business rules and functional terms- Any phase completion - new terms surface naturally as phases complete
Step 0: Current state check
Check for an existing glossary artifact in pureinn-workspace/[project-slug]/artifacts/glossary.md.
Show state table:
| Artifact | Status | Detail |
| Glossary | ✅ / ❌ | [X terms / not yet created] |
If glossary exists: show a summary - how many terms, which domains are covered, when it was last updated.
Interaction: Group related questions (2-4 per round) and confirm before moving on. For any A/B/C/D choice, use the AskUserQuestion tool with one option marked (Recommended) - never print options as plain text. Keep open-ended questions free-text (don't fake options). If the user is unsure, propose 3-4 concrete options plus "Other". Surface an assumption the moment you make one; never fabricate to fill a gap. (Full standard: CLAUDE.md.)
Use AskUserQuestion tool with:
- Question: "What do you want to do in this session?"
- Option A: "Add new terms - paste or describe what to capture (Recommended)"
- Option B: "Extract terms from an existing artifact - paste the artifact text"
- Option C: "Review and update existing entries"
- Option D: "Export full glossary as a clean table or something specific"
If user provides explicit intent with the command (e.g., /pm-glossary add "Feature Set, Delivery Stripe, Stripe"), skip the question and proceed directly.
Step 1: Gather inputs
If adding terms manually (Option A)
Provide the terms you want to add to the glossary.
For each term, share what you know:
- Name (the term, abbreviation, or entity name)
- What it means in this project context
- Which domain it belongs to (Business / Technical / UX / Operations / Legal / Finance / Other)
- Where it is used (which phases, documents, or conversations)
- Any related terms or connected entities
You can provide a raw list and I will structure the entries.If extracting from an artifact (Option B)
Paste the artifact text below.
I will identify terms that should be in the glossary:
- Domain entities and data objects
- Process steps and methodologies
- Artifacts and document names
- Business rules and constraints
- Abbreviations and acronyms
- Roles and stakeholders
- Metrics and KPIs
[paste artifact text]Step 2: Generate or update glossary
Generate in English.
Collision guard (ubiquitous language): if a new term duplicates an existing entry with a different meaning, do NOT silently overwrite and do NOT add a second row with the same name. Surface the conflict to the user - one term carries one meaning per domain, or the shared language breaks down exactly where it matters most. If both senses are genuinely real, qualify the names (e.g. Booking (guest reservation) vs BookingRequest (pre-confirmation)) and cross-link them in Connected Entities. Also check the reverse: two differently-named entries with the same meaning are a synonym pair - keep one canonical, mark the other "synonym of X".
Apply glossary name types consistently:
| Type | Definition | Examples |
| Artifact | A document or structured output produced during the process | PRD, BRD, FSD, Feature Card, Lean Canvas |
| Concept | An abstract idea or principle that shapes how the product is built or defined | PMF, JTBD, MDP, Value Proposition |
| Entity | A domain object: a real-world thing the system tracks or acts on | User, Booking, Feature Set, Stripe, Invoice |
| Metric | A measurement, KPI, or success indicator | NPS, LTV, CAC, AARRR, North Star Metric |
| Process | A methodology, workflow, or repeatable activity | FDD, Design Thinking, KANO Analysis, Sprint |
| Role | A stakeholder, team member, or system actor | PM, Tech Lead, Early Adopter, External API |
| Rule | A business rule, constraint, or governance requirement | Spec Gate, Feature Flag, Backward Compatibility |
| Abbreviation | A short form used throughout documents | B2B, SaaS, POV, HMW, V×C |
ARTIFACT: Project Glossary
# Project Glossary - [Product Name]
> **Phase:** Cross-phase (continuous)
> **Last updated:** [date]
> **Total terms:** [X]
---
## Glossary
| Name | Definition | Domain | Usage Context | Type | Connected Entities | Note |
|---|---|---|---|---|---|---|
| [Term] | [Clear, precise definition in project context] | [Business / Technical / UX / Operations / Legal / Finance] | [Which phases, documents, or conversations use this term] | [Artifact / Concept / Entity / Metric / Process / Role / Rule / Abbreviation] | [Related terms, linked entities, parent/child relationships] | [Disambiguation, edge cases, what this is NOT] |Ordering: Alphabetical within each domain group.
Domain groups (in order):
- Business
- Technical
- UX
- Operations
- Legal / Compliance
- Finance
Internal completeness checklist
<!-- Claude reference only - not shown to user.
Use in Step 0 to identify gaps in existing glossary.
Use in Step 2 to verify output quality before finalizing. -->Each entry must cover:
- Name: the exact term as used in project documents (match casing and spelling)
- Definition: precise, not circular (does not use the term in its own definition)
- Domain: one of the 6 standard domains, or "Other" with explanation
- Usage Context: specific enough to tell someone where to look (e.g., "Used in Feature Card Section 1 and business_rules.md - defines the guard conditions for this entity state transition")
- Type: one of the 8 standard types
- Connected Entities: at least one link where relevant; empty only for standalone abbreviations
- Note: filled when term has a common misuse, overlaps with another term, or is context-specific
Common issues to catch:
- Abbreviations without expansion (e.g., "FDD" without "Feature Driven Development")
- Entities without their parent/child relationships noted (e.g., "Feature" without "Feature Set")
- Process terms without phase assignment in Usage Context
- Rules without the constraint they enforce noted
Notion push
Runs after user approves the glossary artifact.
Read pureinn-variables.md key "Glossary" → get DB URL.
If blank: ask user, save URL to pureinn-variables.md.
Check state.json notion_ids.glossary for cached ID. If not cached: call mcp__claude_ai_Notion__notion-fetch, extract data source ID, save to state.json notion_ids.glossary.
For each term in the glossary, call mcp__claude_ai_Notion__notion-create-pages with both properties AND content. Do NOT use template_id - provide content directly.
properties:
Name: [Term]
Definícia (2–4 vety): [Definition text]
Doména: [Domain]
Typ pojmu: [Type - Entity/Status/Rule/Event/etc]
Kontext použitia: [Usage context]
Prepojené entity/procesy: [Related entities]
content:
## [Term]
[Full definition - 2-4 sentences]
**Domain:** [Domain]
**Type:** [Type]
**Used in:** [Where this term appears in the product]
**Related:** [Related terms or entities]Update if entry with matching Name already exists (use mcp__claude_ai_Notion__notion-update-page).
After push: report counts (created, updated, errors).
Save to
pureinn-workspace/[project-slug]/glossary.mdHandoff
Čo si teraz má: Aktualizovaný glossary - zdieľaný slovník termínov, entít a skratiek, ktorý drží jazyk projektu konzistentný.
Ďalší krok: Pokračuj v aktuálnej fáze - glossary je cross-phase utilita, spúšťaj ho kedykoľvek pribudne nová terminológia. /pureinn ti ukáže ďalší krok podľa fázy.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

