The Agent Skills Directory

[DATA_EXPOSURE]: The skill is specifically designed to locate and analyze sensitive files including .env files, private keys (.pem, .key), and cloud provider configuration files. This activity is the primary intended purpose of the skill (Secrets Management Review). To mitigate risk, the skill includes explicit instructions for the agent to never extract, log, or display actual secret values, only reporting detection patterns and file locations.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests untrusted data from the user's filesystem.
Ingestion points: Files and directories provided via $ARGUMENTS, specifically environment and configuration files analyzed in 'Step 1: Discovery'.
Boundary markers: The skill includes a 'Prompt Injection Safety Notice' that explicitly instructs the agent to treat file content as untrusted data and ignore any instructions found within.
Capability inventory: Uses Read, Grep, and Glob tools for file analysis.
Sanitization: Instructs the agent to report only findings and remediation steps without reproducing the raw data (secrets) from the files.

secrets-management