The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to analyze untrusted source code and configuration files, creating a potential surface for indirect prompt injection attacks.\n
Ingestion points: The agent is instructed to read contents of arbitrary files within a target project directory using shell commands like find and cat, as well as custom scanning scripts (scripts/scan_project.py and scripts/scan_secrets.py).\n
Boundary markers: The skill does not define explicit delimiters or security warnings to prevent the agent from following malicious instructions that might be embedded in the audited code files.\n
Capability inventory: The agent has access to shell command execution and filesystem operations to perform audits and generate reports.\n
Sanitization: The provided scripts use regex patterns to detect issues but do not sanitize the raw file content before the agent processes it.\n- [COMMAND_EXECUTION]: The skill directs the agent to execute various command-line tools to perform security analysis.\n
Evidence: Instructions in SKILL.md include running pip audit, safety check, npm audit, npx auditjs ossi, govulncheck, and trivy. These are industry-standard tools used for their intended security purposes.

skill-security-auditor