The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute git diff and git log commands to retrieve session history and code changes.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it processes untrusted data from the codebase history.
Ingestion points: The skill ingests raw output from git diff and git log commands in SKILL.md (Workflow Phase 1).
Boundary markers: There are no explicit boundary markers or instructions to the model to ignore embedded commands within the ingested git data.
Capability inventory: The skill possesses the Bash, Write, Read, Glob, and Grep tools, which can be leveraged to execute arbitrary commands or modify the filesystem if the model is influenced by the ingested data.
Sanitization: No sanitization, escaping, or validation is performed on the data retrieved from the git history before it is processed by the AI.

insight-extraction