claude-skills / plan-it
CLAUDE.md0 starsChanged 30 days agoMITView on GitHub
Rules bound to /plan-it's own pipeline. Cross-cutting rules (background subagents, model choice, run ledger) live in the root CLAUDE.md. Selection rule (SKILL.md Phase 3): design only for UI/UX surfaces; product only when the feature is new to users or scope is undecided; simplicity only when scope is still open, its cuts are "zur Diskussion" and never auto-applied. The orchestrator runs the git drift check (challengers have no Bash) and passes DRIFT_OUTPUT into each briefing. /plan-it carries effort: high in its…
In this repository13 other files it gives its agents.
- audit/CLAUDE.mdCLAUDE.md
- CLAUDE.mdCLAUDE.md
- delegate/CLAUDE.mdCLAUDE.md
- screens/CLAUDE.mdCLAUDE.md
- ship/CLAUDE.mdCLAUDE.md
- store-assets/CLAUDE.mdCLAUDE.md
- auditSkill
- delegateSkill
- plan-itSkill
- readinessSkill
- screensSkill
- shipSkill
- store-assetsSkill
The file
Files
/plan-it internals
Rules bound to /plan-it's own pipeline. Cross-cutting rules (background subagents, model
choice, run ledger) live in the root CLAUDE.md.
Effort levels
| Level | /plan-it (default high) |
| low | architecture + risk only, no eval |
| medium | architecture + risk always, design/product/simplicity by selection rule, no eval |
| high / xhigh (default) | same selection, full eval |
Selection rule (SKILL.md Phase 3): design only for UI/UX surfaces; product only when the feature is new to users or scope is undecided; simplicity only when scope is still open, its cuts are "zur Diskussion" and never auto-applied. The orchestrator runs the git drift check (challengers have no Bash) and passes DRIFT_OUTPUT into each briefing.
/plan-it carries effort: high in its frontmatter, and per the dual-use rule that value is what
${CLAUDE_EFFORT} receives.
Gotchas
- /plan-it plans are executor-grade handoff artifacts. Template (
plan-it/references/plan-templates.md) stamps the planned-at commit for a drift check, requires a verify criterion per step, machine-checkable done criteria, an out-of-scope list, and STOP conditions, and ends in a## Delegate specsection (/delegate mini-spec format) that/delegate <plan file>reuses; the body before it stays at about 1,500 words./plan-it execute <plan>runs an executor subagent in an isolated worktree and reviews with an APPROVE/REVISE/BLOCK verdict (max 2 revision rounds, merging stays with the user);/plan-it reconcilerefreshes the plan backlog against code drift. Detail:plan-it/references/execute-review.md.
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.