The Agent Skills Directory

PROMPT_INJECTION (LOW): The skill exhibits a vulnerability surface for indirect prompt injection, as several defined personas ingest and analyze external, untrusted data.
Ingestion points: testing_test-results-analyzer.md (parses raw test logs), testing_api-tester.md (processes API responses), and testing_tool-evaluator.md (fetches third-party tool documentation/data).
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the persona definitions.
Capability inventory: The specialists have access to powerful tools including Bash, Write, MultiEdit, and WebFetch.
Sanitization: No data validation or sanitization routines are defined for the ingested content before it is processed by the LLM.
COMMAND_EXECUTION (SAFE): While the skill grants Bash access, all provided command examples (e.g., k6, pytest, ab, dredd) are standard, non-malicious testing utilities used in a transparent manner.

testing-specialist