dash-audit

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill involves command execution in its workflow:
  • It executes a specific local binary at /home/bjorn/.codex/skill-support/bin/ui-audit-preflight to perform a callback map.
  • It directs the agent to verify paths using "repo-native commands" during remediation steps.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing untrusted repository data:
  • Ingestion points: Reads AGENTS.md and other Dash-related source files from the target repository.
  • Boundary markers: No delimiters or instructions are provided to the agent to ignore embedded instructions in the ingested files.
  • Capability inventory: Executes a preflight binary and performs extensive file system reads.
  • Sanitization: There is no mention of sanitizing or validating the input data from the repository before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 02:42 PM
Security Audit — agent-trust-hub — dash-audit