The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references external installation instructions for the infsh CLI tool via a public GitHub repository. This is used to set up the environment required for the skill's functionality.
[COMMAND_EXECUTION]: The skill uses the infsh CLI tool through Bash to perform operations such as logging in, running AI models (e.g., falai/flux-dev-lora), and querying search providers. It also suggests using npx to install supplementary skills from the inference-sh registry.
[PROMPT_INJECTION]: The skill incorporates a surface for indirect prompt injection by processing external data from search assistants.
Ingestion points: External data is ingested through infsh app run tavily/search-assistant and infsh app run exa/answer (as seen in SKILL.md).
Boundary markers: No specific delimiters or boundary instructions are provided to the agent for handling the untrusted search output.
Capability inventory: The skill possesses the capability to execute infsh shell commands via Bash.
Sanitization: No sanitization or validation mechanisms are described for the data retrieved from external search services before it is used to influence agent decisions.

landing-page-design