qa-project-memory

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill functions as a centralized logging system that ingests and persists data from various other AI agent skills and external task outputs into markdown files within the docs/qa-memory/ directory.
  • Ingestion points: The update command and the associated Post-Task Protocol read and process outputs from an ecosystem of QA skills (e.g., test generators, bug ticket creators, flaky detectors).
  • Boundary markers: The skill does not implement specific delimiters or instructions to ignore potential commands embedded within the logs it records.
  • Capability inventory: The skill possesses the capability to write and append to files in the project workspace, creating a persistent 'memory' that is explicitly checked before subsequent tasks.
  • Sanitization: There are no instructions or mechanisms for sanitizing or escaping the text produced by other skills before it is interpolated into the memory files, creating a surface where malicious instructions could be stored and later executed by the agent or other skills reading the log.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 05:38 PM
Security Audit — agent-trust-hub — qa-project-memory