configure-skill-invocation

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes shell commands, specifically recommending rg (ripgrep), to recursively discover skill files within the local and global file systems.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data from SKILL.md files which may contain untrusted instructions.
  • Ingestion points: Reads SKILL.md and agents/openai.yaml files from ~/.agents/skills, .codex/skills, and current project directories.
  • Capability inventory: Includes shell command execution for file discovery and file-writing capabilities to update skill configurations.
  • Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious content within the files it inspects.
  • Sanitization: The skill relies on YAML parsing for metadata inspection but does not specify sanitization or validation logic for the content of the files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 06:33 AM
Security Audit — agent-trust-hub — configure-skill-invocation