gds-retrospective

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python utility script at {project-root}/_bmad/scripts/resolve_customization.py during both the activation phase (Step 1) and the final handoff phase (Step 12) to handle configuration merging and completion tasks.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes content from various project documents which could potentially contain malicious instructions intended to influence the agent's behavior during the retrospective facilitation.
  • Ingestion points: Loads project context, architecture documents, game design documents, story implementation records, and previous retrospective files.
  • Boundary markers: Absent; the workflow lacks explicit delimiters or instructions to treat the loaded document content as untrusted data.
  • Capability inventory: The skill has the ability to execute local Python scripts (via Step 1 and 12) and write/update markdown and YAML files within the project directory (via Step 11).
  • Sanitization: No validation or sanitization is performed on the content of the loaded files before they are summarized or used to drive the interactive team dialogue.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 08:01 PM
Security Audit — agent-trust-hub — gds-retrospective