project-audit

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a consolidated Bash script for project discovery as specified in Dimension 3 and Rule 9. The script is defined as a static template within the skill's own instructions and performs benign, read-only operations such as file existence checks, line counting, and directory listing.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it ingests and processes content from various untrusted project files to perform its audit.
  • Ingestion points: The skill reads contents from CLAUDE.md, ai-context/*.md, openspec/config.yaml, settings.json, analysis-report.md, and local SKILL.md files.
  • Boundary markers: Absent; there are no explicit delimiters or instructions to the agent to disregard malicious instructions that might be embedded within the project files being audited.
  • Capability inventory: The skill uses file reading, globbing, and grep tools. It can write a report to .claude/audit-report.md and execute its internal Bash discovery script.
  • Sanitization: No escaping or validation is performed on the content extracted from the files before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 05:45 AM