tailrocks-reconcile

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell scripts located within the repository, specifically roadmap/<slug>/goal/check.sh, and runs 'done criteria' and gate commands (goal/START.md) parsed from repository files.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes content from the repository, external registries, and web sources that could contain malicious instructions designed to influence agent behavior.
  • Ingestion points: Reads README.md, plan/*.md, and verification/ files within the roadmap/<slug>/ directory.
  • Boundary markers: Instructions explicitly direct the agent to treat repository and web content as evidence only and to flag any embedded instructions.
  • Capability inventory: Includes shell command execution, Git operations (commit, push, move, remove), and writing to markdown files.
  • Sanitization: Relies on the model's ability to identify and disregard instructions embedded in the processed data.
  • [DYNAMIC_EXECUTION]: The skill dynamically executes commands extracted from text-based plan files to verify the status of roadmap items.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 03:12 PM
Security Audit — agent-trust-hub — tailrocks-reconcile