retrospect-report

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local script retrospect-load-sessions.sh using the Bash tool to filter session data based on user-provided arguments.
  • [PROMPT_INJECTION]: The skill extracts content from previous sessions, including user prompts and collaboration insights, to generate aggregate reports. This constitutes an indirect prompt injection surface as it processes untrusted historical data.
  • Ingestion points: Session files and files within the .retro/insights/ directory (SKILL.md).
  • Boundary markers: None identified. Raw session data is interpolated into the report template without delimiters.
  • Capability inventory: Uses Bash, Read, Write, and Grep tools.
  • Sanitization: No validation or escaping of the content extracted from session logs is performed before writing the report.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 06:17 AM
Security Audit — agent-trust-hub — retrospect-report