agentleFS
qyndexqyndex/claude-init/.claude/skills/initiative/SKILL.md

initiative

Skill2 starsChanged 5 months agoMITView on GitHub

Author or amend a multi-quarter initiative — the program-level artifact above per-feature specs. Use for 3+ month efforts spanning multiple specs, multiple teams, or multiple quarters. An 18-month migration is one initiative; a 2-day feature is not. Spec template's `complexity: XL` should split into an initiative + multiple specs.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 1 is distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Rule of thumb: if the spec template's complexity: XL feels right *and* the work spans 2+ quarters, that's an initiative.

    line 27
  2. Distinctive:

    Every initiative links a KR: No KR? Either you don't need this initiative or your OKRs are wrong.

    line 52
  3. Very specific:

    Update roadmap.md to reflect.

    line 62
  4. Very specific:

    Write a .claude/memory/decisions/ entry capturing the program-level lessons.

    line 63
  5. Very specific:

    Update OKRs.md — close out KR if applicable.

    line 64

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Initiative

The artifact above the spec. Bind multi-quarter programs into a coherent thread.

When to create an initiative vs. a spec

SignalSpecInitiative
Effortdays–weeksquarter–year+
Owners1-2 engineers3+ engineers, sometimes multiple teams
Phasesmaybe (in plan.md)required (multiple specs, gated)
KR alignment1 KR1+ KRs
Code reachbounded featureplatform-wide
External dependenciesrarecommon (legal, vendor, regulatory)
Decision cadenceoncemonthly review

Rule of thumb: if the spec template's complexity: XL feels right and the work spans 2+ quarters, that's an initiative. Split it.

Process

ALL initiatives use the [[grill-me]] skill for discovery — the decisions cascade. Picking a KR shapes phase breakdown shapes flag namespace shapes spec catalog. Batched questions force premature commitment. Round 5 A2.

  1. Invoke grill-me with branch order: KR alignment → success metric → phase 0 → phase 1+ → flag namespace → risk register. One question at a time, recommendation + reasoning per question.
  2. Define phases — each phase ships independently. P0/P1/P2/...; each phase has an exit criterion. Output: filled ## Phases table.
  3. Catalog specs — list specs that already exist + specs needed. Output: filled ## Spec catalog.
  4. Risk register — top 5 risks, mitigation, owner. Output: filled ## Risks.
  5. Flag policy — flag namespace, owner, default cleanup window. Output: filled ## Feature flag policy.
  6. Save — initiatives/active/<id>-<slug>.md from initiatives/templates/initiative.md.
  7. Cross-link — update roadmap.md, link KRs in OKRs.md.

In AUTOPILOT context (Cloud Routine / --bg): grill-me uses recommendations as defaults and surfaces [OQ-pending-operator-review] in the handoff. Initiative status becomes draft-autopilot-needs-review, not approved.

Commands

  • /initiative create <slug> — grill-me-driven authoring (one-at-a-time, recommendation per question)
  • /initiative status <id> — phase progress, % done, risk health
  • /initiative close <id> — mark shipped/abandoned, archive
  • /initiative supersede <id> --by <new-id> — link supersession

Hard rules

  • Every initiative links a KR. No KR? Either you don't need this initiative or your OKRs are wrong. Don't paper over.
  • Phases must ship independently. "Phase 1: scaffolding" is not shippable. Rewrite.
  • Risks have owners. A risk without an owner is a wish.
  • Flag namespace is declared. Without it, flags collide across initiatives.
  • Spec catalog stays current. Architect agent updates it when authoring/closing a spec under the initiative.

When an initiative closes

  1. Mark status: shipped (or abandoned / superseded).
  2. Move initiatives/active/<file> → initiatives/archive/.
  3. Update roadmap.md to reflect.
  4. Write a .claude/memory/decisions/ entry capturing the program-level lessons.
  5. Update OKRs.md — close out KR if applicable.

References

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.