The Agent Skills Directory

[PERSISTENCE]: The skill uses a cronjob to schedule a weekly task that runs a local Python script. This behavior allows the skill to maintain its functionality across sessions through automated background execution.
[COMMAND_EXECUTION]: The skill instructions direct the agent to use the terminal tool to run administrative commands, specifically executing a catalog refresh script and invoking the hermes CLI for fallback chain verification.
[EXTERNAL_DOWNLOADS]: The skill connects to https://opencode.ai/zen/v1/models to retrieve the latest list of available models. This network activity is necessary for the core functionality of the skill but involves fetching data from an external provider.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from an external API (OpenCode) to update the agent's model catalog.
Ingestion points: The model catalog is fetched from https://opencode.ai/zen/v1/models via scripts/refresh_opencode_free_catalog.py.
Boundary markers: None present in the instructions to delimit or validate the incoming model IDs.
Capability inventory: The skill uses the terminal tool for subprocess calls and CLI operations.
Sanitization: There is no explicit mention of sanitizing or validating the model names or metadata fetched from the remote source before they are processed by the agent.

opencode-free-rotation