The Agent Skills Directory

[DATA_EXFILTRATION]: The Python script embedded in the skill performs network operations via the requests library to test tool performance against dynamic endpoints.
Evidence: The _test_performance method in SKILL.md uses requests.get(api_endpoint, timeout=10) to measure response times.- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it ingests untrusted configuration data and uses it to perform network operations without sanitization.
Ingestion points: The tool_config dictionary passed to the evaluate_tool method in SKILL.md.
Boundary markers: No delimiters or instructions are provided to the agent to ignore potentially malicious embedded content within the configuration data.
Capability inventory: The skill has the capability to perform network requests (requests.get) as defined in SKILL.md.
Sanitization: There is no evidence of validation or sanitization for the api_endpoint or other fields within the tool_config object.

agency-tool-evaluator