The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected in the instructions or code examples.
[COMMAND_EXECUTION]: The skill provides standard curl command examples to demonstrate how to interact with the FairStack API. This is expected behavior for a technical documentation skill.
[DATA_EXPOSURE]: The documentation demonstrates safe practices by instructing users to use environment variables for API keys and providing placeholder values (fs_live_...) instead of hardcoded credentials.
[PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection, as it transmits user-supplied text to an external API to receive recommendations. This is the primary function of the skill and is handled following standard integration patterns.
Ingestion points: User-provided descriptions are ingested via the prompt parameter in the POST /v1/select-model endpoint (SKILL.md).
Boundary markers: None explicitly defined in the skill documentation.
Capability inventory: Performs network POST operations to https://fairstack.ai (SKILL.md).
Sanitization: No sanitization or escaping logic is described in the integration examples.

fairstack-smart-select