agentleFS
Sign inStart free
zdnemzzdnemz/claude-team/.claude/skills/continue-work/SKILL.md

continue-work

Skill1 starsChanged 4 months agoMITView on GitHub

Autonomous execution loop — pick the next ready tasks from the plan, dispatch specialist agents with the right model, verify results, record progress, commit. Use when state phase is PLANNED or EXECUTING.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. rahulrajaram/rahulskills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository5 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 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.

  • 3

    Very specific

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

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

    If phase is PLANNED: state_set {phase:"EXECUTING", detail:"<current epic>"}.

    line 8
  2. Very specific:

    Not yet registered this session → spawn general-purpose with the body of .claude/agents/<role>.md prepended to the prompt + the task's …

    line 19
  3. Very specific:

    Run the task's verify step yourself.

    line 24
  4. Distinctive:

    Keep looping while ready tasks remain and context is healthy.

    line 34

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Continue Work

If phase is PLANNED: state_set {phase:"EXECUTING", detail:"<current epic>"}. Then loop:

1 — Load

progress_status → if a handoff note exists (memory_get context/handoff), apply it then memory_delete context/handoff. Read docs/plans/legend.md; pick the current epic (first not-done whose dependencies are done); read its epic file. For architecture/spec context, query the knowledge graph (/graphify) instead of re-reading whole spec files.

2 — Pick

Ready tasks = status todo with all depends done. Prefer finishing an in-flight story over opening a new one. Up to 3 tasks may run as parallel subagents ONLY if they touch disjoint files; otherwise sequential.

3 — Dispatch

For each task, spawn its role agent with its model (both named in the task line):

  • Named agent type available → use it.
  • Not yet registered this session → spawn general-purpose with the body of .claude/agents/<role>.md prepended to the prompt + the task's model param.

The dispatch prompt MUST contain: task id + title, the story's acceptance criteria, deliverable, verify step, paths of relevant specs, and: "Report files changed + verification command + output."

4 — Verify (you, not them)

Run the task's verify step yourself. Story complete → spawn qa-engineer for a story-level check against its acceptance criteria. Fail → one retry with the failure context. Fail again → escalate model one tier (registry escalation rule). Still failing → mark task blocked, memory_save blockers/<task-id> (symptom, attempts, what's needed), move on if other tasks are workable.

5 — Record (never skip, every task)

  • Tick the task checkbox in the epic file (todo→done, or mark blocked).
  • progress_log {status:"done", task:"T-…", role:"…", note:"…"}.
  • Decisions made along the way → memory_save decisions/<slug>.
  • Git repo exists → commit per completed task: T-NNMMKK: <title>.

6 — Continue or stop

Keep looping while ready tasks remain and context is healthy. Stop and summarize when:

  • A story/epic milestone lands → report, ask user whether to continue.
  • A blocker needs the user → surface it plainly. state_set BLOCKED only if NOTHING else is workable.
  • Context nearly exhausted → memory_save context/handoff (in-flight work, exact next step), progress_log, tell the user: type start in a fresh session to resume.
  • Last epic done → state_set {phase:"DONE"}, final report.

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.

Check your setup, free