ork-verify
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to identify and execute canonical commands found within the repository's instructions to validate readiness and run tests.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from repository files and external library documentation to determine execution parameters and verification logic.
- Ingestion points: Repository instructions and library documentation retrieved via the
context7MCP server are used to define which commands the agent runs. - Boundary markers: No explicit delimiters or boundary markers are specified to wrap untrusted repository data when it is processed by the agent.
- Capability inventory: The skill allows for the execution of shell commands for testing and verification as defined in the target repository.
- Sanitization: No specific sanitization or validation of the repository-provided commands is mandated, though the instructions do require the agent to confirm flags and semantics using a documentation tool.
Audit Metadata