Install
$ agentstack add skill-zaixincheng174-ai-codex-agent-governance-skills-issue-slice-drafter ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
issue-slice-drafter
Source repo: https://github.com/mattpocock/skills Source path: skills/engineering/to-issues/SKILL.md Source commit: b8be62ffacb0118fa3eaa29a0923c87c8c11985c Local adaptation note: Adapted from an issue-publishing workflow into a read-only drafting skill. It keeps vertical tracer-bullet slicing, HITL/AFK classification, dependencies, and acceptance criteria, while removing automatic issue tracker writes.
Purpose
Turn an approved plan or clarified goal into draft slices that are independently understandable and verifiable.
Each slice should be a thin vertical tracer bullet through the relevant layers, not a horizontal task such as "build backend" or "write UI".
Activation
Use when the user asks to break down work into issues, slices, implementation tickets, agent-ready tasks, or a next-action board.
This skill may also be used after a Goal Contract or audit finds a clear implementation backlog, but only as a drafting step.
Slice Requirements
For each draft slice include:
Title:
Type: HITL or AFK
Parent/source:
What to build:
Acceptance criteria:
Dependencies:
Verification:
Out of scope:
Evidence references:
Classification:
- HITL: requires human decision, product judgment, design review, credential access, approval, or high-risk interpretation.
- AFK: can be implemented and verified from available repo context and explicit acceptance criteria.
Workflow
- Gather source material from the current conversation, Goal Contract, plan, PRD, audit finding, or referenced file.
- If repo evidence is needed to avoid vague slices, inspect the relevant code/docs first.
- Draft the smallest useful set of vertical slices.
- Mark dependencies explicitly.
- Include acceptance criteria and verification commands or evidence expectations where known.
- Ask for approval before any issue tracker action.
Forbidden actions
- Do not publish GitHub issues, GitLab issues, Linear tickets, labels, comments, PRs, branches, commits, or pushes without explicit approval.
- Do not run
gh issue createor equivalent issue tracker commands unless the user explicitly approves that exact action. - Do not apply labels or mark anything ready-for-agent unless the user explicitly approves.
- Do not modify parent issues or source documents.
- Do not edit AGENTS, update CONTEXT.md, create ADR, deploy, git push, or run setup.
- Do not turn a narrow Goal Contract into a broad backlog.
Stop conditions
Stop when:
- The source goal is not clear enough to slice safely.
- The requested output would require publishing rather than drafting and approval is absent.
- A slice cannot be made independently verifiable.
- The work is single-step and an issue breakdown would add noise.
Output Pattern
Issue Slice Drafts
Source:
Assumptions:
Slices:
1.
Type:
Dependencies:
Acceptance criteria:
Verification:
Publishing status: draft only; no issue tracker writes performed
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: zaixincheng174-ai
- Source: zaixincheng174-ai/codex-agent-governance-skills
- License: MIT
- Homepage: https://github.com/zaixincheng174-ai/codex-agent-governance-skills/blob/main/docs/demo-false-pass.md
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.