agentleFS
simscalsimscal/sapiens-dtj/.claude/skills/user-stories/SKILL.md

user-stories

Skill0 starsChanged 4 months agoMITView on GitHub

Use when writing, reviewing, splitting, or refining user stories and acceptance criteria for a product backlog. Owns story + AC quality — the INVEST gate, the per-AC testability linter, and the structured story-spec contract. Touches no filesystem or external system; the caller owns all orchestration.

25 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use brainstorm-experiments-new, brainstorm-experiments-existing and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • skytiger6724Skill

    brainstorm-experiments-new

    Design lean startup experiments (pretotypes) for a new product. Creates XYZ hypotheses and suggests low-effort validation methods like landing pages, explainer videos, and pre-orders. Use when validating a new product idea, creating pretotypes, or testing market demand.

    Also in 9 projects that use this

  • skytiger6724Skill

    brainstorm-experiments-existing

    Design experiments to test assumptions for an existing product — prototypes, A/B tests, spikes, and other low-effort validation methods. Use when validating assumptions, testing feature ideas cheaply, or planning product experiments.

    Also in 7 projects that use this

  • 45ckSkill

    pestle-analysis

    Also in 7 projects that use this

  • About-IntelligenceSkill

    positioning-ideas

    Brainstorm product positioning ideas differentiated from competitors.

    Also in 7 projects that use this

  • launchapp-devSkill

    interview-script

    Create a structured user interview script with warm-up, core exploration, and wrap-up sections. Use when preparing for user research interviews to ensure consistent, insightful conversations.

    Also in 10 projects that use this

  • About-IntelligenceSkill

    product-name

    Brainstorm 5 unique, memorable product names with rationale aligned to brand values and target audience.

    Also in 7 projects that use this

  • +6 more from skytiger6724

The file

Show its rules and text

We split this skill into the rules it gives an agent

13 rules. 0 are common, 11 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 24 edited versions we read keep these.

  • 11

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every story satisfies all six (first three checked during drafting, last three during validation):

    line 52
  2. Distinctive:

    Run on every AC; each must pass all four:

    line 74
  3. Very specific:

    Rewrite: when the requirement is clear but the wording weak.

    line 83
  4. Distinctive:

    Escalate: when the requirement itself is unclear (missing trigger, ambiguous outcome, unknown threshold) — re-open Discovery.

    line 84
  5. Distinctive:

    Every Discovery capability covered by ≥1 story's ACs.

    line 92
  6. Distinctive:

    No implementation details as AC outcomes.

    line 94
  7. Distinctive:

    No foundational work as stories.

    line 95
  8. Very specific:

    Every cross-story dependency explicit in depends_on_titles.

    line 96
  9. Distinctive:

    No gold-plating — every AC traces to stated intent or a surfaced edge case.

    line 97
  10. Distinctive:

    Never invent scope — run Discovery.

    line 112
  11. Distinctive:

    Never add technical details to ACs.

    line 113
  12. Distinctive:

    Never silently rewrite an AC whose underlying requirement is unclear — escalate.

    line 114
  13. Distinctive:

    Never touch GitHub, filesystem, or external systems.

    line 115

“Common” is measured: kept by at least 6 in 10 of the 24 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

User Story Expertise

Discovery → draft via the Story Spec → Testability Linter on every AC → Validation → Ordering → emit per Notes Rendering.

Discovery Protocol

Synthesise understanding across the six categories; any gap that would change scope is blocking. Probe each:

  1. Role — who specifically, with context (not "user").
  2. Intent — the outcome in user terms, not implementation.
  3. Value — why it matters; if only statable as tautology, not ready.
  4. Definition of done — the observable success condition.
  5. Constraints — platform, compliance, performance, permissions, flows to preserve.
  6. Edge and failure cases — invalid input, missing permission, network failure, missing data, repeated action.

Present synthesis → ask every blocking gap via AskUserQuestion (max 4 per call; batch if more) — concrete options per question, never open-ended prose; don't draft yet. Re-synthesise per response until no blocking gaps remain → confirm in one line before output. Non-blocking ambiguities → resolve with a stated assumption.

Scope Boundary

User stories capture user-observable behaviour only. Out of scope as stories: design tokens, primitives, shared components, schemas, migrations, data models, infrastructure, cross-cutting modules, refactors, spikes.

Foundational work needed -> never smuggle it in or invent a "technical story":

  • Team prerequisite → separate task/spike (outside this skill); add to depends_on_titles only if it blocks the user-observable outcome.
  • User asked for technical breakdown → stop; say this skill produces user stories only; offer to list technical work as plain items.

Story Spec

Draft each story to this spec — Scope Boundary governs what becomes a story, INVEST (first three) and AC Format govern content, Splitting when too big:

- title: "[Story] <title>"
  user_story: "As a <role>, I want <action> so that <benefit>"
  acceptance_criteria:
    - "<AC>"
  notes:
    edge_cases: ["<text>", ...]
    design_instructions: [{label: "<filename.md>", url: "<url>"}, ...]
    mockups: [{label: "<filename.html>", url: "<url>"}, ...]
    references: [{label: "<text>", url: "<url>"}, ...]
    depends_on_titles: ["<title>", ...]

INVEST

Every story satisfies all six (first three checked during drafting, last three during validation):

  • Independent — deliverable without another story first; unavoidable dependencies go in depends_on_titles.
  • Negotiable — the "what" is fixed, the "how" open; no implementation in the want clause.
  • Valuable — observable value to a user or stakeholder; a tautological "so that" means re-run Discovery on intent.
  • Estimable — scope clear enough to ballpark; fails on vague scope, unknown dependencies, hidden research.
  • Small — completable within a sprint by one team (see Splitting).
  • Testable — every AC passes the linter.

AC Format

  • When X, then Y. or Given X, when Y, then Z.
  • Checklist bullets when the story is about completeness, not behaviour. Don't force Given/When/Then where it doesn't fit.
  • An AC may cite a reference by label instead of restating detail.
  • Target 2–5 ACs; fewer suggests under-specification, more suggests splitting.

Splitting

Story failing Small or with >~5 ACs splits. Techniques, preference order: workflow steps → happy path vs alternates → CRUD operations → data/input variation → interface surface. Never split by technical layer (frontend/backend story) — violates Independent + Valuable.

Testability Linter

Run on every AC; each must pass all four:

  1. Observable — verifiable without reading code or internal state.
  2. Located — names the user-visible area/screen/surface.
  3. Specific — concrete trigger and outcome, not "works correctly".
  4. Verifiable — a non-engineer can confirm pass/fail in a running system.

On failure choose one:

  • Rewrite when the requirement is clear but the wording weak. Preserve meaning; record {original, rewritten, reason}, aggregate as rewrote_for_testability.
  • Escalate when the requirement itself is unclear (missing trigger, ambiguous outcome, unknown threshold) — re-open Discovery; never silently rewrite.

Cap: more than three rewrites for one story → stop and escalate the whole story.

Validation

Across the full set:

  • Every Discovery capability covered by ≥1 story's ACs.
  • No duplicate scope.
  • No implementation details as AC outcomes.
  • No foundational work as stories.
  • Every cross-story dependency explicit in depends_on_titles.
  • No gold-plating — every AC traces to stated intent or a surfaced edge case.

Ordering

Dependency depth first (no depends_on_titles first), then user value within each depth level.

Notes Rendering

  • design_instructions — - Design Instructions: [label](url). Multiple → nested list.
  • mockups — same pattern under - Mock UI:.
  • edge_cases — one bullet per item.
  • depends_on_titles — - Depends on: <resolved refs>.

Constraints

  • Never invent scope — run Discovery.
  • Never add technical details to ACs.
  • Never silently rewrite an AC whose underlying requirement is unclear — escalate.
  • Never touch GitHub, filesystem, or external systems.
  • AC reshape: only ## Acceptance Criteria and ## Notes may change.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.