The Agent Skills Directory

[COMMAND_EXECUTION]: The skill validates traceability metadata by executing a local Python script at plugins/testany-eng/scripts/trace_lint.py. This script is identified as a tool provided by the vendor, testany-io.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external Business Requirements Documents (BRDs) to extract user journey data, representing an attack surface where malicious instructions could be embedded in the source text. 1. Ingestion points: The agent reads user-provided BRD files during the initial context loading phase. 2. Boundary markers: No specific delimiters or security instructions are implemented to isolate the untrusted document content from the agent's core instructions. 3. Capability inventory: The skill has the ability to write local documents and execute internal shell commands. 4. Sanitization: No sanitization or validation of the input text from the BRDs is described before interpolation into the prompt.

uc-interviewer