The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs shell-based git operations to discover and revert commits.
Evidence: Executes git log, git status, and git revert commands.
Evidence: Interpolates variables such as {trackId}, {X.Y}, and {sha} directly into shell command templates.
[INDIRECT_PROMPT_INJECTION]: The skill processes project-specific data files to determine task status and commit ranges, which represents a potential surface for indirect injection if these files are manipulated by an external actor.
Ingestion points: conductor/tracks.md, plan.md, and metadata.json (SKILL.md).
Boundary markers: None specified for file content parsing.
Capability inventory: File system read/write access and subprocess execution via git commands.
Sanitization: No explicit sanitization or validation of the content within the project files before processing.

conductor-revert