The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local CLI commands including entire status and entire checkpoint explain to retrieve repository metadata and historical intent transcripts. It also uses git log to identify relevant checkpoints within commit trailers.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data originating from Entire checkpoints and git commit messages. This creates a surface for indirect prompt injection where maliciously crafted transcripts could attempt to influence the agent's behavior.
Ingestion points: Output from entire checkpoint explain and git log is read into the agent's context (SKILL.md).
Boundary markers: The instructions mandate a 'Transparency Principle', requiring the agent to explicitly label findings as 'recorded intent' (from checkpoints) or 'inferred from code' to separate data from reasoning.
Capability inventory: The skill utilizes entire CLI commands and git for data retrieval; it does not perform arbitrary file writes or unsafe code execution based on the ingested data.
Sanitization: The skill requests structured JSON output (--json) from the CLI tool to ensure consistent parsing and reduce the risk of interpretation errors.

using-entire