agentic-actions-audito…
Audit GitHub Actions that run AI agents for prompt injection, unsafe interpolation, sandbox gaps, and permissive actor rules. Use…
Turn one confirmed security finding into a disclosure-ready GitHub advisory with root cause, proof of concept, impact, and source evidence. Use for reporting an established vulnerability, not discovering or validating one.
$ npx -y skills add waybarrios/opencode-power-pack --skill vuln-report --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/vuln-reportContext preview
The summary Claude sees to decide when to auto-load this skill.
Turn one confirmed security finding into a disclosure-ready GitHub advisory with root cause, proof of concept, impact, and source evidence. Use for reporting an established vulnerability, not discovering or validating one.
name: vuln-report description: "Turn one confirmed security finding into a disclosure-ready GitHub advisory with root cause, proof of concept, impact, and source evidence. Use for reporting an established vulnerability, not discovering or validating one." license: MIT (modified; see UPSTREAMS.json)
Draft one disclosure-ready report for one confirmed bug. Keep the report evidence-driven, concrete, and concise. Prefer the section order and phrasing rules in [references/report-template.md](references/report-template.md).
1. Confirm the report is about one bug only. 2. Extract the minimum facts needed to prove the issue:
3. Separate demonstrated facts from inference. State assumptions explicitly. 4. Draft the report using the required section order from [references/report-template.md](references/report-template.md). 5. Always embed at least one fenced code snippet from the decisive code path, and explain what each snippet proves. 6. Always convert repository file references and patch references into GitHub markdown links, and prefer embedding those links directly into the surrounding explanation instead of listing them separately. 7. Add only the optional sections that materially improve accuracy or triage value. 8. Save the final report as `report.md` inside a folder named with the bug's severity identifier (`C1`, `H1`, `M1`, etc.) followed by a lowercase hyphenated slug derived from the final report title. Use `C` for Critical, `H` for High, `M` for Medium, sequentially numbered if there are multiple bugs of the same severity. Example: `C1-cross-site-websocket-hijacking-re-enabled-by-allow-websocket/report.md`. Also, ensure the bug report title and internal references use this ID (e.g., '[C1] Cross-Site WebSocket Hijacking'). Do not write reports for Low severity findings — document them in the summary table only. 9. Remove filler, hedging, and unproven claims before finalizing.
Always include these sections in this order:
1. `Summary` 2. `Details` 3. `Root Cause` 4. `Proof of Concept (PoC)` 5. `Impact`
If the repository already uses `Technical Details` with `Root Cause` nested under it, preserve that local pattern. Otherwise keep `Root Cause` as its own section.
`report.md` is a disclosure-ready artefact. The reader must understand the vulnerability, the trace, the impact, and the reproduction without opening any sibling working file (drafts, debate transcripts, review notes, internal metadata).
Open with the vulnerable behavior, attacker control, and outcome in one short paragraph. Name the component only if it improves clarity.
Explain the code path and why the protection fails. Include relevant conditions such as auth mode, stateless mode, parser behavior, MIME confusion, or transport assumptions. Support the explanation with code snippets and GitHub markdown links to the exact source locations.
State the design or implementation mistake in one focused subsection. Prefer causal language such as missing origin validation, unsafe trust in extension-derived MIME, or policy enforced only in one execution mode.
Use the shortest reliable reproduction. Prefer numbered steps and a runnable request, command, or code block. State the expected result.
Describe exploitability and consequence, not just severity labels. Cover who is exposed, what an attacker gains, and which environments are most at risk.
Include an optional section only when it adds concrete triage value.
Allowed optional sections include:
54 rigorous skills for Codex, OpenCode, and Pi: code review, security audit, feature development, frontend design, MCP tools, Hugging Face ML/training, and more.
Audit GitHub Actions that run AI agents for prompt injection, unsafe interpolation, sandbox gaps, and permissive actor rules. Use…
Audit and improve project-rules files (AGENTS.md, CLAUDE.md, .agents/instructions, local overrides) so the agent keeps accurate…
Capture learnings from the current session into the project-rules file (AGENTS.md, CLAUDE.md, or local override) so future…
Operational rubric that turns "don't make AI slop" into observable properties, severity levels, evidence requirements, and repair…
Design a feature architecture by analyzing existing codebase patterns and conventions, then provide a comprehensive…
Deeply analyze an existing codebase feature by tracing execution paths, mapping architecture layers, understanding patterns and…