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.
The required workflow reads the ticket body from a GitHub issue via `scripts/stamp.sh` and the PR body and diff of a GitHub pull request via `/review-pr`, and anyone can open an issue or PR on GitHub without the agent first selecting it, so this is medium.
agranado2k/agentic-sdlc
0f5671d
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.