implement
Implement a piece of work based on a spec or set of tickets. Use when the user says "implement", or asks to build, execute or work on a ticket, spec or plan.
- License 许可
- MIT
- 所属仓库
- meldomapp/skills
- 最近更新
- 2026-11-14
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- implement
- description
- Implement a piece of work based on a spec or set of tickets. Use when the user says "implement", or asks to build, execute or work on a ticket, spec or plan.
02 资源文件 1
- 参考文档agents/openai.yaml96 B
03 SKILL.md 正文
Implement the work described by the user in the spec or tickets.
If the user passes a ticket reference, fetch it from the issue tracker and state its title before starting. If the reference is ambiguous, ask.
Call the Skill tool with "meldom:tdd" where possible, at pre-agreed seams.
Run typechecking regularly, single test files regularly, and the full test suite once at the end.
Once done, call the Skill tool with "meldom:code-review" to review the work.
Commit your work to the current branch.
On Meldom
- Review the whole run, then commit. Capture
git rev-parse HEADbefore building. After all tickets supplied to this run are implemented, run the final checks and callmeldom:code-reviewagainst that captured commit, covering every ticket together. Fix findings, rerun affected checks and review until there are no findings. Then commit throughmeldom:ship. - Claim before building. Move the ticket you are about to build, and its parent spec, to
in_progresswithmcp__meldom__ticket_batch_update, keyed by each ticket's ULID (theidfield), never itsKEY-42key. Work on one child ticket at a time; its parent staysin_progressthrough final review. - Report progress. On the first pass, send one plan with exactly one step per ticket, each its ticket's key and sized
S,MorL:mcp__meldom__progress({ "goal": "<spec title>", "plan": ["KEY-42:M"] }). Moving a ticket todonemarks its step done, so send nodonefor it. Send{ "add": ["KEY-43:S"] }for each ticket added later, and{ "finish": true }with the run's last tool call. If the tool is not there after one search, skip every progress call silently. - Mark each ticket done as it passes. The moment a ticket's acceptance criteria pass,
mcp__meldom__ticket_updateit todonewith areason, then claim the next one. - Build what you file. Work the build surfaces beyond the tickets is a follow-up: file it at once with
mcp__meldom__ticket_create,parent_idset to the spec or the ticket's own parent, and build it in this session. Before you finish, readmcp__meldom__conversation_status; while any ticket this session created is open, build it. The one exception is a step only a human can take: hand it over throughmeldom:wizardor a question. - Record the final review per ticket. For each ticket the review's Spec axis confirms, set
mcp__meldom__ticket_outcome({ "id": <ulid>, "outcome": "verified" }). A ticket it contradicts gets"failed", goes back toopen, and is built again. - Close the parent. Parent status never rolls up on its own: after final checks and review pass and every child is
doneorclosed, move the parent todonewithreason: "All children done: <summary>", then check its own parent the same way.
04 同仓库更多 Skills
writing-for-agents
Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.
wizard
Generate an interactive bash wizard that walks a human through steps only they can perform. Use when provisioning infrastructure, setting up credentials or CI secrets, walking an unfamiliar third-party dashboard, or running a one-off migration or cutover. Don't invoke this for steps the agent can perform itself.
wayfinder
Plan a huge chunk of work (more than one agent session can hold) as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.
wait-what
Stop. That last message did not land: re-pitch it.
triage
Move issues and external PRs through a state machine of triage roles, categorise, verify, grill if needed, and write agent-ready briefs.
to-tickets
Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker (edges as text in one file per ticket locally, or native blocking links on a real tracker).