skills.rest
bfly123
bfly123COMMUNITY
Explore all skills by this author →
GitHub

task-detail-packet

Drafts task-scoped detail artifacts and a JSON detail packet for orchestrator workflow import.

3.5k|342|Updated Oct 25, 2025
Search Tags:#acceptance criteria#agentic workflow# orchestration#json manifest#Multi-Agent#task detailing#planner replan
One-click install
npx skills add https://github.com/bfly123/claude_code_bridge --skill task-detail-packet
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: task-detail-packet
Source: https://github.com/bfly123/claude_code_bridge/tree/main/docs/plantree/plans/agentic-loop-workflow/drafts/agentroles.ccb_task_detailer/skills/task-detail-packet
Command: npx skills add https://github.com/bfly123/claude_code_bridge --skill task-detail-packet

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Macro task packets in an agentic loop workflow often lack the task-local execution detail needed before an orchestrator can dispatch workers. This Skill fills that gap by drafting task-scoped detail design, acceptance criteria, and a machine-readable detail packet that a script-owned workflow can import directly.

Core Features & Use Cases

  • Task Detail Drafting: Reads controller-supplied macro task refs, plan-tree refs, accepted decisions, source files, tests, and evidence to produce a task-scoped detail design and source-evidence map.
  • Structured Detail Packet: Returns task-detail-design.md, brief-update-summary.md, and exactly one canonical detail-packet.manifest.json section conforming to the ccb.detail_packet_manifest.v1 schema.
  • Outcome Classification: Classifies results as local_detail_ready, planner_replan_required, needs_clarification, or blocked, with global impact rated none, bounded, or macro.
  • Use Case: During an agentic loop workflow, a controller activates this Skill on a macro task packet; the Skill returns a detail packet the orchestrator imports to decide whether to dispatch workers or request a planner replan.

Quick Start

Ask the agent to draft a task detail packet for the current macro task packet using the controller-supplied refs and evidence, returning the manifest JSON in the reply.

Frequently Asked Questions about task-detail-packet

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a task detail packet for an agentic workflow?▼

Activate the Skill with controller-supplied macro task refs, plan-tree refs, accepted decisions, source files, tests, and evidence. It returns task-detail-design.md, brief-update-summary.md, and one detail-packet.manifest.json section in the reply for script-owned import.

What schema does the detail packet manifest use?▼

The manifest uses schema ccb.detail_packet_manifest.v1 and contains exactly four fields: schema, detail_result, readiness, and global_impact. Only specific legal tuples are allowed, such as local_detail_ready/detail_ready/none.

When does the task detailer request a planner replan?▼

A planner replan is requested only when the result is classified as planner_replan_required. In that case the Skill authors a versioned ccb.detailer.replan_request.v1 and submits exactly one silent inline ask to the resident planner through the managed capability.

Can this Skill dispatch workers or run CCB commands?▼

No. The Skill never dispatches workers, runs ccb plan, ccb loop, ccb ask, or wrapper commands, and never edits authoritative CCB state. The only routing exception is the single managed planner replan handoff.

Where are the detail artifacts written?▼

Artifacts are not written to the project tree. The detail design, brief update summary, and manifest JSON are returned as reply content so the owning script can import them, preventing self-import loops.

Related Skills

Explore More Software Engineering Skills→
lunaticwithaducklunaticwithaduck

product

Generate product planning artifacts like PRDs, roadmaps, and user stories.

Community
Advanced
paulmalmquistpaulmalmquist

winston-router

Route Winston/Novendor requests to correct agents and handoffs using surface rules.

Community
Advanced
wojonswojons
4

data-validation

Validate data quality, types, schemas, and consistency across systems using Python.

Community
Advanced
Jtapias92672Jtapias92672

mayor-worker-pattern

Coordinate multiple autonomous agents with a Mayor-Worker hierarchy and ledger-based task handoffs.

Community
Advanced