From dev-spec-kit
The completion ritual — run when a feature's tasks are done and it's time to land the branch. Fresh-evidence entry gate, a fixed option menu, typed confirmation for destructive paths, provenance-checked cleanup, every step journaled. Use when the user says finish/land/merge/wrap up, or after dev-spec-kit pr is ready.
How this skill is triggered — by the user, by Claude, or both
Slash command
/dev-spec-kit:dev-spec-kit-finishThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Entry is EARNED, the menu is FIXED, destruction is TYPED, and every step lands in the journal.
Entry is EARNED, the menu is FIXED, destruction is TYPED, and every step lands in the journal.
Run, in order, and SHOW the output (after any session break: dev-spec-kit resume FIRST):
dev-spec-kit graph build — MUST exit 0 (every proof green at the current tree). Stale/red → run
dev-spec-kit drift, fix, and re-enter. Old green output does not count; the evidence must be fresh.dev-spec-kit verify — MUST exit 0 (Build ALL + every configured kind green, journaled on the
current tree). The PR gate refuses without it.dev-spec-kit status — all tasks for this feature DONE.dev-spec-kit pr --title "<title>" — body generated, gate verdict green.
If any step fails: STOP. Report what blocked. Do not present the menu.dev-spec-kit pr --create).discard exactly. Anything else aborts.Do not add commentary, recommendations, or a fifth option. Wait for the choice.
.worktrees/ or .claude/worktrees/ — provenance check: never clean up a workspace dev-spec-kit did not
create.dev-spec-kit approve <taskIds> --note "<the human's words>" if the human gate wasn't recorded yet.Every executed step is already audit-logged (cli.run); close with dev-spec-kit log -n 10 so the human
sees the recorded trail: graph green → approval → PR/merge → cleanup. The ritual is complete only
when the journal shows it.
npx claudepluginhub pratiyush/dev-spec-kitProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.