session-memory

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes external data from repository files (e.g., README.md, CLAUDE.md) and user arguments. This creates a surface for indirect prompt injection. Ingestion points: Phase 1 and Phase 3 read project files and $ARGUMENTS. Boundary markers: Absent. Capability inventory: Local file writes and git command execution. Sanitization: Absent. The skill's structured formatting requirements provide partial mitigation against malicious content execution.
  • [PROMPT_INJECTION]: The instructions 'Do not ask for confirmation. Read, persist, and report' encourage the agent to execute documentation tasks autonomously, which reduces user oversight for file modifications.
  • [COMMAND_EXECUTION]: The skill invokes 'git log --oneline -20' and 'git diff' to audit project changes when no arguments are provided. These commands are executed locally within the project directory to support session logging.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 10:36 AM
Security Audit — agent-trust-hub — session-memory