The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes source code from the active repository, which is considered untrusted data.
Ingestion points: Files are read and searched using Read, Glob, Grep, and ast-grep tool executions in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the ingested file content.
Capability inventory: The skill possesses the Bash, Write, Read, Glob, and Grep capabilities.
Sanitization: The instructions include guidance on escaping metavariables in shell commands and manually spot-checking results for accuracy.
[COMMAND_EXECUTION]: The workflow relies on executing the ast-grep CLI via the Bash tool. The agent dynamically constructs these commands based on user input, though the skill provides instructions in SKILL.md and references/search-recipes.md to ensure correct escaping and testing.

ast-grep