memory-manage

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted project content to update the agent's persistent memory layer.
  • Ingestion points: The skill reads from project configuration files, READMEs, source files, tests, and CI/CD configurations during the init and update modes.
  • Boundary markers: No explicit delimiters or instructions are used to separate untrusted content from the agent's core instructions.
  • Capability inventory: Performs broad file system read operations and writes structured documentation to the ai-context/ directory.
  • Sanitization: No validation, escaping, or filtering is applied to the data retrieved from project files before it is used to generate or update the documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 05:45 AM