ov-agentic-workflows
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and follow natural language protocols defined in workflow source files (.github/workflows/*.md). This allows external content to influence the agent's behavior during workflow modification tasks.
- Ingestion points: Workflow source files (.github/workflows/*.md) and shared prompt fragments.
- Boundary markers: The skill advises following the workflow's own instructions and shared protocols without specifying protective delimiters.
- Capability inventory: Compilation and execution of shell commands via 'gh' and script execution via 'python3'.
- Sanitization: No explicit sanitization or instruction filtering is mentioned.
- [COMMAND_EXECUTION]: The skill utilizes several command-line operations for development tasks.
- Evidence: Uses 'gh aw compile' for workflow generation and 'python3 -c' for JSON schema validation.
- [EXTERNAL_DOWNLOADS]: The skill manages the installation of the 'gh-aw' extension from a well-known source.
- Evidence: Recommends installing or upgrading the 'gh-aw' extension from the 'github/gh-aw' repository.
Audit Metadata