matthew-a-carr/draft-epic
Draft an EPIC from a GitHub issue and open a PR for review. Use when triggered by a routine on `Issue opened` with label `ai:plan-epic`, or when a user asks to "draft an epic from issue #NNN". Non-interactive — proceeds on best interpretation and surfaces unresolved questions in the EPIC's §Open Questions section rather than blocking. The PR review loop resolves ambiguity. Does NOT write child SPECs — slice SPECs are drafted lazily by the `draft-spec` routine when the human files one `ai:plan` issue per slice after the EPIC PR is merged.
83
90%
Does it follow best practices?
Impact
87%
1.14xAverage score across 2 eval scenarios
Low
Low-risk findings worth noting
Low
Low-risk findings.
1 low severity finding. Worth noting, but not necessarily harmful.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
Routine mode reads the GitHub issue body via `mcp__github__issue_read` (outsider-authored issue/PR/comment text from other contributors) and then uses that text to draft the EPIC and PR body, so it can feed arbitrary free-form outsider content into the LLM context.
- Audited
- Security analysis