verify-openspec-docs

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill explicitly instructs a subagent to "Re-run the terminal commands shown" within documentation units. This behavior introduces a risk where malicious shell commands embedded in documentation files could be executed by the agent. The skill includes a mitigation suggestion to use "read-only commands" or "scratch directories," but the underlying capability depends on the agent's ability to execute arbitrary commands found in user-controlled content.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes documentation prose as external, untrusted data to guide agent behavior.
  • Ingestion points: Documentation files identified by <PAGE PATH> and <DOCS TREE ROOT>/README.md.
  • Boundary markers: Absent; the subagent reads the documentation content directly without escaping or delimiters to separate instructions from data.
  • Capability inventory: The subagent can read files, execute shell commands, and run the openspec validate utility.
  • Sanitization: No sanitization, filtering, or validation is performed on the documentation content before it is parsed and acted upon by the subagent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 09:30 PM
Security Audit — agent-trust-hub — verify-openspec-docs