The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to execute the local searchapi binary to interact with the SearchAPI.io service. It also utilizes shell utilities like jq for processing JSON results, which is a standard pattern for CLI-based tools.
[REMOTE_CODE_EXECUTION]: The documentation provides instructions to install the CLI tool by cloning the official repository from GitHub (github.com/agoodway/searchapi) and running a build command (just install). This represents the standard installation process for the vendor's own software.
[EXTERNAL_DOWNLOADS]: The CLI tool performs network requests to SearchAPI.io to retrieve search data. Additionally, the fetch-images utility allows for downloading images from external URLs to the local file system as part of its core functionality.
[INDIRECT_PROMPT_INJECTION]: The skill acts as an ingestion point for untrusted data from external search results (Google, Zillow, etc.), which could theoretically contain malicious prompts intended to influence the agent.
Ingestion points: Search results returned in JSON or table format by the searchapi binary, as described in SKILL.md.
Boundary markers: Absent; the skill does not explicitly instruct the agent to use delimiters or ignore instructions found within search result data.
Capability inventory: The skill possesses the ability to execute shell commands (searchapi, jq) and perform file system writes (fetch-images).
Sanitization: Absent; the skill relies on the agent's internal reasoning to handle external content.
[SAFE]: The skill implements proactive security measures by instructing the agent to mask API keys and avoid dumping the ~/.searchapi.json configuration file into the chat context.

searchapi