agile-v-core

Pass

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to ingest and act upon data from local files within the .agile-v/ directory, specifically REQUIREMENTS.md, POLICY.yaml, and CONTROL_MATRIX.yaml. This creates a surface for indirect prompt injection where instructions embedded in project requirements could influence the agent's behavior during synthesis or verification phases.
  • Ingestion points: Files located in the .agile-v/ directory, including requirements, policies, and control matrices.
  • Boundary markers: The skill does not define explicit delimiters or instructions to ignore embedded prompts within these data sources.
  • Capability inventory: The skill allows the agent to synthesize artifacts and write to various log files (e.g., TRACE_LOG.md, DECISION_LOG.md, CHECKPOINTS.md).
  • Sanitization: No explicit sanitization or validation of external requirement content is mentioned prior to processing.
  • [EXTERNAL_DOWNLOADS]: The frontmatter metadata references a GitHub repository (https://github.com/gsd-build/get-shit-done) as a source for its context engineering concepts. This is documented as a conceptual reference rather than a runtime code dependency.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 13, 2026, 06:07 AM
Security Audit — agent-trust-hub — agile-v-core