agent-harness-design
Designing agent harnesses and tool systems — risk taxonomy for tools, permission decisions, draft/commit pattern, structured tool…
Use when a large plan, PRD, feature, refactor, research plan, or multi-step coding task must be split into small ready-for-agent tickets with acceptance criteria, verification commands, blockers, and vertical tracer-bullet slices. Do not use for small tasks that should be
$ npx -y skills add AnastasiyaW/codex-claude-code-config --skill plan-to-tickets --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/plan-to-ticketsContext preview
The summary Claude sees to decide when to auto-load this skill.
Use when a large plan, PRD, feature, refactor, research plan, or multi-step coding task must be split into small ready-for-agent tickets with acceptance criteria, verification commands, blockers, and vertical tracer-bullet slices. Do not use for small tasks that should be
name: plan-to-tickets description: Use when a large plan, PRD, feature, refactor, research plan, or multi-step coding task must be split into small ready-for-agent tickets with acceptance criteria, verification commands, blockers, and vertical tracer-bullet slices. Do not use for small tasks that should be implemented directly, single-bug fixes, or chat-only summaries.
Turn a large plan into small, independently executable tickets. Use this before handing a broad plan to agents, before long autonomous work, or when the user asks for tickets/issues/slices.
If no issue tracker is explicitly selected, write local tickets under:
`<project>/.agent/tickets/<YYYY-MM-DD>-<slug>/`
Each ticket is one Markdown file:
`TICKET-001-short-slug.md`
Each ticket must include these headings exactly:
`## Status` must contain `ready-for-agent` only when all acceptance criteria and verification steps are concrete.
`## Acceptance Criteria` must contain checklist items (`- [ ] ...`).
`## Verification` must contain at least one command, script, test, artifact check, or explicit manual-review gate.
1. Read the plan/spec and current repo context needed to avoid fake tickets. 2. Draft ticket files in dependency order. 3. Run the validator:
python "$env:USERPROFILE\.claude\claude-code-config\scripts\validate_agent_tickets.py" --tickets-dir <ticket-dir>
4. Fix every validator failure before claiming the ticket split is ready. 5. Report the ticket directory and the count of ready tickets.
A practical configuration kit for Claude Code, Codex, and other coding agents. It contains architectural principles, enforcement hooks, skills, drop-in rules, starter templates, and dynamic-workflow commands.
Designing agent harnesses and tool systems — risk taxonomy for tools, permission decisions, draft/commit pattern, structured tool…
Use when designing, auditing, refactoring, or explaining an agentic harness for any domain, especially when work must continue…
Практическая инженерия диффузионных моделей: архитектуры, обучение, инференс, оптимизация памяти. Использовать при любых задачах…
Expert prompt engineering for FLUX.2 [klein] image generation and editing model. Use this skill whenever the user wants to create…
Plan or review LoRA and edit-training work specifically for FLUX.2 Klein or Qwen-Image-Edit, including paired datasets,…
Forensic image-to-prompt compiler for image generation models. Use this skill whenever the user wants to: convert/describe an…