The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from repository files during the scanning process. This creates an attack surface where malicious files could attempt to influence the agent's behavior through embedded instructions. The risk is mitigated by the skill's lack of network access and the constrained nature of its reporting task.
[COMMAND_EXECUTION]: The skill executes local shell scripts and system tools such as git, grep, and file. These operations are performed within the local repository context to identify potential security risks and are consistent with the skill's stated purpose.
[SAFE]: The skill serves a legitimate security purpose and contains no indicators of malicious intent. It does not perform unauthorized network operations, exfiltrate data, or utilize obfuscation techniques.

safe-repo