opencode-free-rotation
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEPERSISTENCECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PERSISTENCE]: The skill utilizes the cronjob tool to schedule a weekly task that refreshes the model catalog. This is a transparently disclosed use of persistence mechanisms for the skill's primary stated purpose.
- [COMMAND_EXECUTION]: The skill uses the terminal tool to execute local Python scripts and CLI commands within the repository, such as scripts/refresh_opencode_free_catalog.py and hermes fallback list, to maintain configuration state.
- [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The skill dynamically fetches a list of available models from an external API at https://opencode.ai/zen/v1/models in SKILL.md.
- Boundary markers: No explicit delimiters or ignore-instructions are mentioned in the fetch process to protect against malicious instructions embedded in the external JSON catalog.
- Capability inventory: The skill has access to terminal shell execution and persistence scheduling through the cronjob tool.
- Sanitization: The documentation does not specify any validation, schema enforcement, or sanitization of the fetched external JSON data before it is processed by the agent.
Audit Metadata