The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for using standard CLI tools such as git and gh (GitHub CLI). It specifically mentions the git bisect run command, which automates testing by executing a local script (e.g., ./scripts/check-feature-x.sh). This is a legitimate development workflow utilizing local repository assets.
[INDIRECT_PROMPT_INJECTION]: The archaeology workflow involves the agent reading external, potentially unvetted data from git commit logs, pull request descriptions, and issue trackers. This constitutes a surface for indirect prompt injection.
Ingestion points: Instructions in SKILL.md direct the agent to read output from git log, gh pr list, and gh issue list.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the history retrieval steps.
Capability inventory: The skill utilizes subprocess execution for git and gh CLI operations.
Sanitization: There is no specified sanitization or filtering of the retrieved git history or issue text.

blame-archaeology