The Agent Skills Directory

[SAFE]: The skill defines efficient, read-only search patterns using standard shell utilities like find and grep for code exploration. It explicitly prohibits any file modifications or write operations, ensuring the integrity of the codebase.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it reads and processes content from external codebase files which are untrusted. However, the associated risk is negligible as the skill lacks capabilities for network access, file system modification, or arbitrary command execution that could be exploited by malicious content within those files.
Ingestion points: Codebase files accessed via grep and read commands as described in SKILL.md.
Boundary markers: No specific delimiters or instructions to ignore embedded instructions are defined for the processed data, though the skill body itself contains strong restrictive instructions to remain read-only.
Capability inventory: The skill is strictly limited to read-only operations; it explicitly forbids write operations such as git commands, npm install, or directory creation.
Sanitization: No sanitization or validation of the ingested file content is performed.

lightweight-explorer