Skip to main content
ClaudeWave
Skill3.5k repo starsupdated 3d ago

task-detail-packet

Draft task-scoped detail artifacts and a detail packet as reply content for script-owned workflow import.

Install in Claude Code
Copy
git clone --depth 1 https://github.com/SeemSeam/claude_codex_bridge /tmp/task-detail-packet && cp -r /tmp/task-detail-packet/docs/plantree/plans/agentic-loop-workflow/drafts/agentroles.ccb_task_detailer/skills/task-detail-packet ~/.claude/skills/task-detail-packet
Then start a new Claude Code session; the skill loads automatically.

SKILL.md

# Task Detail Packet

Use this skill when a macro task packet needs task-local execution detail before
orchestrator dispatch.

## Workflow

1. Read only the controller-supplied macro task refs, plan-tree refs, accepted
   decisions, source files, tests, and durable evidence for this immaculate
   activation.
2. Draft task-scoped detail design and source-evidence map.
3. Produce detailed acceptance, verification, and worker handoff notes.
4. Return `task-detail-design.md`, `brief-update-summary.md`, and exactly one
   canonical `detail-packet.manifest.json:` section immediately followed by a
   literal `json` fence. The packet is JSON, never Markdown.
5. Classify the result as exactly `local_detail_ready`,
   `planner_replan_required`, `needs_clarification`, or `blocked`.
6. In the brief update, classify `global impact: none|bounded|macro`, give a
   compact rationale, and state planner backfill evidence.
7. For `planner_replan_required` only, author the versioned request and submit
   exactly one direct silent inline ask to resident `planner` through the
   managed capability. For every other result, submit no Planner ask.

The manifest must contain exactly `schema`, `detail_result`, `readiness`, and
`global_impact`. Set `schema` to `ccb.detail_packet_manifest.v1`. Legal tuples
are `local_detail_ready/detail_ready/none`,
`planner_replan_required/planner_replan_required/macro`,
`needs_clarification/needs_clarification/none|bounded`, and
`blocked/blocked/none|bounded|macro`. Do not use an old label, a schema-named
fence, a second manifest, or a cross-outcome tuple.

## Boundaries

- Do not rewrite roadmap or accepted decisions directly.
- Do not lower acceptance criteria.
- Do not dispatch runtime agents.
- Never dispatch workers or submit downstream asks except the one restricted
  `ccb.detailer.replan_request.v1` Planner replan handoff.
- Do not directly edit authoritative CCB state or runtime files.
- Do not run `ccb plan`, `ccb loop`, generic `ccb ask`, `ccb_test`, or wrapper
  commands. The sole managed Planner handoff is the only routing exception.
- Do not write detail artifacts into the project tree for later self-import;
  put artifact content in the reply.
- Do not add `--chain`, wait, watch, poll, use arbitrary targets, or run a
  generic shell/CCB command.
- Provider and model selection remain project configuration concerns.
ccb-githubSkill

Maintain this CCB project's GitHub-facing release and npm publication surface. Use when preparing, publishing, auditing, or fixing CCB releases; updating README.md, README/zh.md, localized README files, CHANGELOG.md, VERSION, package.json, GitHub release notes/assets, repository description/topics, npm registry state, or GitHub Actions release/test status.

decisionsSkill
ccb-configSkill

Private built-in CCB configuration skill for agentroles.ccb_self. Design, edit, validate, and prepare reloads for .ccb/ccb.config, role bindings, providers, windows, workspaces, tool windows, sidebar, and provider startup inputs. Use only inside ccb_self; non-self agents should delegate CCB config changes to ccb_self.

ccb-self-chainSkill

Diagnose and repair CCB ask/job/message/reply/artifact/callback lineage. Use for missing replies, incomplete artifacts, pending callbacks, retry/resubmit/ack decisions, reply delivery problems, or work-chain resume advice.

ccb-self-diagnoseSkill

Diagnose CCB runtime, mounted daemon graph, tmux namespace and panes, provider context, queue/inbox/trace, replies/artifacts, config drift, and storage boundaries. Use when the user asks what is broken, which agent is stuck, whether CCB is mounted, why a reply did not arrive, or what to check first.

ccb-self-recoverSkill

Recover CCB agents, panes, mounts, provider contexts, API/provider failures, config reload aftermath, clear operations, and guarded single-agent restarts. Use when the user asks to fix, recover, restart if safe, clear context, reload, remount, or keep work going after provider/API failure.

askSkill

Send a request to a CCB agent with `ask`.

ccb-clearSkill

Clear CCB managed agent conversation context with `ccb clear`. Use when the user writes `$ccb-clear`, `$ccb_clear`, or asks to clear/reset one or more CCB agent contexts without restarting or deleting project state.