agentleFS
LiriRaidLiriRaid/waymark/skills/dept-product/SKILL.md

dept-product

Skill2 starsChanged 6 days agoMITView on GitHub

Product. FIRST when an idea must become scoped work: \"quiero un feature de…\", \"necesito que la app haga…\", \"historia de usuario\", \"criterios de aceptación\".

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 7 are distinctive, and 2 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.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Done: when — the DoD items that apply, including preserved behavior for change requests?

    line 28
  2. Very specific:

    Does not own: technical design and placement → dept-architecture · UI/visual decisions → dept-ux-ui · test implementation → dept-qa · …

    line 32
  3. Distinctive:

    Write an explicit out-of-scope list for every L2+ task.

    line 46
  4. Very specific:

    Use the Plan agent or plan mode before coding at L3 and for any change touching shared/core layers.

    line 48
  5. Distinctive:

    Prefer: express requirements in the domain language used by the codebase (entity and feature names).

    line 50
  6. Distinctive:

    Prefer: include the non-functional constraints the stack profile flags (rendering mode, platform targets).

    line 51
  7. Distinctive:

    Prefer: keep each task independently verifiable and mergeable.

    line 52
  8. Distinctive:

    Never plan speculative abstractions or features for a hypothetical future.

    line 53
  9. Distinctive:

    Never jump to code when the change touches more than 3 files without a written plan.

    line 54

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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.

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

Product Management

Quick ref

Mission: Turn a request into verifiable requirements, acceptance criteria and a scoped breakdown before code is written. Rules: Given/When/Then criteria for every L2+ change · explicit out-of-scope · smallest scope that delivers the value · plan before coding at L3 Skills by default: Explore · Plan · library-docs DoD: Requirements unambiguous and testable, criteria mapped to tests, scope and out-of-scope written.

Entry

Guest (no Waymark block in context) → ../waymark/references/coexistence.md §3. Reads:

  • Learnings: ~/.waymark/learnings/dept-product.md if it exists.
  • Stack profile: L1 Commands, L2+ full.

Brief questions

The brief must answer:

  1. What is being delivered, in one sentence (actor, capability, outcome)?
  2. Why / for whom — which user or role gains what value?
  3. Where does it land — which feature/module per the architecture, and which departments implement it?
  4. What is out of scope for this iteration?
  5. Which unknowns block the outcome (ask) and which are assumed (state them)?
  6. How is it verified — the acceptance criteria and the test or manual step for each?
  7. How — procedure from this skill, level, and the skills/agents used (Explore, Plan, library-docs)?
  8. Done when — the DoD items that apply, including preserved behavior for change requests?

Scope

  • Owns: requirement elicitation, clarification, acceptance criteria, scope and out-of-scope, task breakdown, prioritization, Definition of Done per task.
  • Does not own: technical design and placement → dept-architecture · UI/visual decisions → dept-ux-ui · test implementation → dept-qa · delivery pipeline → dept-devops.

Procedures

In procedures.md (same folder), one section per task:

  • New feature (L2/L3)
  • Bug report
  • Change request on existing behavior
  • Scope triage (request is too large)
  • New project from scratch
  • Ask vs assume

Rules

  • Have written acceptance criteria in Given/When/Then for every L2+ task before the first edit.
  • Write an explicit out-of-scope list for every L2+ task.
  • Convert relative dates ("next sprint", "tomorrow") to absolute dates when writing plans or saving memory.
  • Use the Plan agent or plan mode before coding at L3 and for any change touching shared/core layers.
  • Escalate the level as soon as the breakdown exceeds the current level's scope.
  • Prefer: express requirements in the domain language used by the codebase (entity and feature names).
  • Prefer: include the non-functional constraints the stack profile flags (rendering mode, platform targets).
  • Prefer: keep each task independently verifiable and mergeable.
  • Never plan speculative abstractions or features for a hypothetical future.
  • Never jump to code when the change touches more than 3 files without a written plan.

Tools

CapabilitySkill / MCP / AgentWhenLevel
memoryWaymark memory: the pack the hook hands over, W memory / tasks, and engram search (or mem_search)Prior decisions and root causes; what you learn goes in the Cierre's Aprendido (the hook saves it to engram)L1
search.codebaseExplore agentScoping needs more than 3 locationsL1
plan.implementationPlan agent (or plan mode when approval is needed)Multi-layer features, shared/core changes, L3 breakdownsL3
docs.librarylibrary-docs (→ the docs MCP servers the user has)Feasibility depends on an unverified library/framework APIQ
app.runrunObserve current behavior before a change requestL2
requirements.authoringnone yet → waymark references/skills.mdStory mapping / backlog tooling—

Definition of Done

  • Requirements numbered, singular, verifiable; assumptions listed
  • Acceptance criteria in Given/When/Then cover happy, error and boundary paths
  • Out-of-scope list written
  • Every criterion mapped to a test or a manual verification step
  • Preserved behavior listed for change requests
  • Breakdown tagged with department skill and level
  • Non-obvious scope decisions in the Cierre's Aprendido (the hook saves them to engram)

Hand-offs

  • To dept-architecture: placement, boundaries, new modules, new projects, or any L3 decision.
  • To dept-ux-ui: flows, states, copy and visual requirements for UI work.
  • To dept-frontend / dept-backend: implementation of each task.
  • To dept-data: new entities, schema changes, persistence or caching requirements.
  • To dept-security: requirements touching auth, permissions, secrets or personal data.
  • To dept-qa: acceptance criteria to turn into tests.
  • To dept-devops: release, environment or rendering-mode constraints.

Learned rules

Grows with use (../waymark/references/learning.md).

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.