The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The atoms-search.py script makes network requests to api.xyskill.xyz to retrieve business 'atoms' (principles and evidence) if the local database is unavailable. This domain is consistent with the vendor's naming conventions and infrastructure.
[COMMAND_EXECUTION]: The skill uses the python3 command to execute a local script atoms-search.py. This is a standard method for implementing tool-assisted reasoning in AI skills.
[INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it retrieves data from external sources (API responses and local .jsonl files) and interpolates them into the agent's context. However, these sources are intended to be curated business knowledge pieces provided by the developer.
Ingestion points: Search results retrieved from references/atoms.jsonl or the api.xyskill.xyz search endpoint, and user-provided business details in SKILL.md Phase 1B.
Boundary markers: The skill does not explicitly use delimiters to wrap the retrieved 'atom' content before it is processed by the AI.
Capability inventory: Execution of a local Python script; network requests via urllib.request; file system read access for configuration and database files.
Sanitization: The atoms-search.py script performs keyword matching and scoring but does not include explicit sanitization of the retrieved text strings.

xy-biz-scan