The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to install the tool locally using pip install . and then execute the web-search-plus command. This pattern involves executing code from the repository and encourages the agent to modify the environment's PATH variable if the executable is not immediately found.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it retrieves and processes content from external search providers (Serper, Tavily, Exa, etc.) which is then parsed by the agent to answer user queries.
Ingestion points: Content is ingested from search provider results, specifically the results, answer, and metadata fields as defined in references/output-schema.md.
Boundary markers: Absent; the instructions in SKILL.md do not provide delimiters or specific guidelines for the agent to ignore or isolate instructions potentially embedded in the search results.
Capability inventory: The agent has the capability to execute shell commands (web-search-plus) and access the local file system (via .env and config.json files) as described in SKILL.md and references/env.md.
Sanitization: There is no mention of sanitization, escaping, or validation of the data retrieved from external web sources before it is processed by the agent.

web-search-plus-cli