which-skill
Route a fuzzy request to the right skill in this library. Use when the user is unsure which skill fits, asks 'which skill should I use for X', describes a task without naming a skill, or when a request could plausibly match several skills. Produces a best-fit recommendation with the inputs to gather, a runner-up with the tie-breaker, and a workflow recipe when the job spans multiple skills.
git clone --depth 1 https://github.com/mohitagw15856/pm-claude-skills /tmp/which-skill && cp -r /tmp/which-skill/exports/obsidian/pm-essentials/which-skill/which- ~/.claude/skills/which-skillwhich-skill.md
# Which Skill Router
Given a fuzzy professional ask ("my boss wants an update on the Q3 launch"), pick the single best skill in this library to run — and say why — instead of making the user browse 400+ options.
## What This Skill Produces
- The **best-fit skill** for the request, with a one-line justification
- The **inputs to gather** before running it (from that skill's Required Inputs)
- A **runner-up skill** and the tie-breaker that separates them
- A **workflow recipe** recommendation instead, when the job genuinely spans 3+ skills
## Required Inputs
Ask for (if not already provided):
- **The task in the user's own words** (even one sentence is enough)
- **Who the output is for** (audience changes the pick: a board deck is not a team update)
- **One-off or recurring?** (a monitor/briefing skill differs from a one-time analysis)
## Routing Method
1. **Name the artifact.** What lands on someone's desk when this is done — a PRD, a ranked list, a briefing, a plan? Route on the deliverable, not on topic keywords.
2. **Search the catalog — never route from memory.** Read `SKILLS.md` (the auto-generated listing grouped by domain), or search with `npx pm-claude-skills list` / the MCP `search_skills` tool. Match the user's phrasing against skill `description` trigger phrases.
3. **Prefer the specific skill over the general one.** A skill built for the exact artifact (e.g. `ab-test-readout` for analysing a finished test) beats a broader neighbour (`experiment-designer`).
4. **Check the disambiguation table below** for the known look-alike clusters before answering.
5. **Escalate to a workflow recipe** (see `WORKFLOWS.md`, e.g. `/ship-a-feature`, `/launch-a-product`) when the ask needs 3+ chained skills — don't recommend the skills one by one.
6. **Recommend, don't interrogate.** Ask at most one clarifying question, and only when the answer would change the pick.
## Disambiguation Table — look-alike clusters
| You want… | Use | Not |
|---|---|---|
| A one-off deep teardown of a rival (SWOT, positioning map) | `competitor-teardown` | `competitive-analysis` |
| A full landscape doc: feature matrix, win/loss, battlecard inputs | `competitive-analysis` | `competitor-teardown` |
| A recurring "what changed in the market this week/month" briefing | `competitive-intelligence-monitor` | `competitor-signal-tracker` |
| A read on one specific competitor announcement | `competitor-signal-tracker` | `competitive-intelligence-monitor` |
| Release notes straight from a raw git log / commit list | `changelog-generator` | `changelog-writer` |
| A Keep-a-Changelog entry from an already-curated change list | `changelog-writer` | `changelog-generator` |
| Positioning, messaging pillars, use cases — the GTM *content* | `go-to-market` | `go-to-market-planner` |
| A tiered launch plan with cross-functional coordination — the GTM *operation* | `go-to-market-planner` | `go-to-market` |
| Themes from interview transcripts specifically | `user-interview-synthesis` | `user-research-synthesis` |
| Synthesis across mixed sources (surveys, feedback, transcripts) | `user-research-synthesis` | `user-interview-synthesis` |
| Pure RICE scoring of a backlog | `rice-prioritisation` | `feature-prioritisation` |
| Choosing/applying a framework (RICE, MoSCoW, Kano, ICE) | `feature-prioritisation` | `rice-prioritisation` |
| RICE blended with strategic-fit weighting | `rice-impact-matrix` | `rice-prioritisation` |
| A summary *of an existing document* for executives | `executive-summary` | `executive-update` |
| A standalone product briefing *written for* the C-suite | `executive-update` | `executive-summary` |
| A BLUF-style project status update for stakeholders | `stakeholder-update` | `executive-update` |
| Designing an experiment before it runs (sample size, guardrails) | `ab-test-planner` | `ab-test-readout` |
| Analysing a finished test and making the ship/no-ship call | `ab-test-readout` | `ab-test-planner` |
## Output Format
### Skill Recommendation
**Best fit:** `skill-name` — [one line: why this artifact matches the ask]
**Before you run it, have ready:**
- [input 1 from that skill's Required Inputs]
- [input 2]
**Runner-up:** `other-skill` — pick this instead if [the tie-breaker condition].
**Run it:** `/skill-name` in Claude Code, or open it in the [Playground](https://mohitagw15856.github.io/pm-claude-skills/).
*(If a workflow fits better)* **This is a multi-skill job** — run `/recipe-name` (chains `a` → `b` → `c`), because [why the chain beats a single skill].
## Quality Checks
- [ ] The pick was verified against the live catalog (SKILLS.md / search), not recalled from memory
- [ ] Every look-alike cluster the ask touches was checked against the disambiguation table
- [ ] The recommendation names the concrete artifact the user will get, not a topic
- [ ] The runner-up includes a real tie-breaker condition, not "also good"
- [ ] Multi-skill jobs point to one workflow recipe, not a list of 4 skills to run manually
## Anti-Patterns
- [ ] Do not recommend more than two skills — a router that returns a list has not routed
- [ ] Do not route on topic keywords ("competitor" ≠ always `competitive-analysis`); route on the deliverable
- [ ] Do not ask a chain of clarifying questions — one at most, and only if it changes the pick
- [ ] Do not invent skill names — if nothing in the catalog fits, say so and suggest `SKILL_REQUEST.md`
- [ ] Do not recommend a general skill when a specific one exists for the exact artifact
---
<!-- Run as an AI-plugin prompt. {{selection}} is the Text Generator / Templater
variable for the highlighted text — replace it with your plugin's equivalent
(e.g. {} in Copilot for Obsidian), or paste your input there manually. -->
Apply the skill above to the following input:
{{selection}}Design a 360-degree feedback survey or write a structured 360 feedback report. Use when asked to build a 360 feedback process, write 360 feedback for a colleague, design a feedback survey, or produce a feedback report. Produces either a complete survey instrument with rating scales and open-ended questions, or a structured narrative feedback report with themes, strengths, and development areas.
Decode a 401k or workplace retirement plan — the real cost of its funds, the match's fine print, vesting math, and the plan features worth using or avoiding. Use when someone asks 'is my 401k any good', 'decode my 401k plan', 'which funds should I look at', or 'what fees am I paying'. Produces a fee decode in dollars-over-time, match and vesting math, a fund-lineup triage by cost, and the questions for HR or the plan administrator.
Design statistically rigorous A/B tests for product features, UI changes, onboarding flows, and pricing experiments. Use when asked to set up an experiment, design an A/B test, calculate sample size, or interpret test results. Produces a complete test plan with hypothesis, variant definitions, sample size, duration estimate, guardrail metrics, and a results interpretation guide.
Analyse a finished A/B test and write the readout — the result, whether it's statistically and practically significant, what it means, and the ship/no-ship call. Use when asked to analyse experiment results, write an A/B test readout, interpret test data, or decide whether to ship a variant. Produces a clear verdict with the lift and confidence, segment cuts, the risks (peeking, novelty, sample), and a recommendation. Distinct from planning a test — this reads results.
Generate a WCAG 2.2 accessibility audit checklist and remediation suggestions for any UI or design. Use when asked to audit for accessibility, check WCAG compliance, review a design for a11y issues, or create an accessibility remediation plan. Produces a prioritised checklist with pass/fail assessments and specific fixes.
Plan a trip that actually works with a disability or access need — confirm real accessibility (not just 'accessible' labels), book the assistance in advance, plan for equipment and medication, and build in the contingencies for when access breaks down. Use when someone says 'plan an accessible trip', 'travelling with a wheelchair/disability', 'book assistance for my flight', or 'will this hotel actually work for me'. Produces an access-verified itinerary, an assistance-booking checklist, an equipment/medication plan, and contingency scripts. Verify specifics with providers.
Request a reasonable accommodation at work or in education — frame it around the barrier and the adjustment (not your diagnosis), cite the right process, and navigate the back-and-forth constructively. Use when someone says 'I need a workplace accommodation', 'request reasonable adjustments', 'ADA/Equality Act accommodation', or 'how do I ask for accommodations for my disability/condition'. Produces the request letter, a barriers-and-adjustments map, disclosure guidance, and a plan for the interactive process. Not legal advice — routes to the formal process and to advocacy where needed.
Build a structured account plan for any key customer or target account. Use when asked to create an account plan, key account strategy, strategic account review, or territory plan. Produces a complete account plan with relationship map, growth opportunities, risks, and 90-day action plan.