aiskyhub/codex_with_cc/codex-with-cc-planning

Claude Code and agent skill by aiskyhub/codex_with_cc

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Plan codex-with-cc workflows by turning a subagent or delegation request into scoped tasks, roles, acceptance criteria, and review gates before dispatch.

Install

npx vskill@latest install aiskyhub/codex_with_cc/codex-with-cc-planning

SKILL.md

Codex with CC Planning

Read ../codex-with-cc/CODEX_WITH_CC.md before planning. Use this skill when a request needs child-agent, subagent, delegation, 子代理, 委派, or 派工 routing through codex-with-cc.

Produce a concise workflow design before any dispatch. The output must be usable as TaskFile content:

  • Assign one WorkflowId for the user request.
  • Split work into task-file-sized assignments with stable TaskId values and explicit dependencies recorded as -DependsOn.
  • Choose one role per task: planner, implementer, researcher, reviewer, or final-verifier.
  • Define Scope, allowed writes, forbidden writes, verification commands, and a stable SessionKey for each task.
  • Mark each task as serial, parallel read-only, or parallel writable with non-overlapping scope.
  • Define acceptance criteria and both required review gates for every implementer task: spec review first, then quality review.
  • Include a final-verifier task for every workflow with implementer tasks; workflow verification rejects accepted implementation work without it.
  • Give every task Goal, Allowed Scope, Forbidden Actions, Acceptance Criteria, Verification, and Report Requirements.
  • Plan a validate_delegate_task.* pre-dispatch check for non-trivial task files and any task with -Tests or reviewer metadata.
  • Use ../codex-with-cc/contract.json values for roles, status tokens, and report headings; do not invent local variants.

Planning quality checks:

  • Do not dispatch a task that depends on an unresolved product decision.
  • Do not parallelize tasks that may write the same file or mutate the same behavior.
  • Treat implementer tasks as writable; parallel implementers must have explicit non-overlapping scopes.
  • Prefer a small researcher task before implementation when the code path is unknown.
  • Do not plan inline -Task, legacy -Mode, or implicit session-key fallback.
  • Return NEEDS_CONTEXT when safe decomposition is not possible from the current request and repository context.

Do not run claude or delegate_to_claude.* from the main thread.

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
193
Repository forks
26
7-day trend score
26

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/aiskyhub/codex_with_cc/codex-with-cc-planning/badge)](https://verified-skill.com/skills/aiskyhub/codex_with_cc/codex-with-cc-planning)

Something wrong with this skill or its page? Tell us.