The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and summarize untrusted external data from version control systems and user inputs, which could potentially contain malicious instructions aimed at influencing the agent's behavior.
Ingestion points: The skill reads data from git commit logs (--from-git mode) and Pull Request lists/descriptions (--from-prs mode) to identify changes.
Boundary markers: The instructions do not define specific delimiters or "ignore instructions" markers to isolate the content of the commits/PRs from the skill's processing logic.
Capability inventory: The skill has access to Bash, Grep, Glob, Read, Write, and Edit tools, primarily used for reading repository history and writing the final markdown documents.
Sanitization: No specific sanitization or validation logic is defined to filter out potentially malicious natural language instructions embedded within commit messages or PR bodies.

release-notes