openspec-apply-change

Pass

Audited by Gen Agent Trust Hub on Jul 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill follows security best practices by utilizing the allowed-tools frontmatter field to restrict the agent's environment. This limits the Bash tool to only executing commands prefixed with openspec:, preventing arbitrary command execution.
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to interact with the local environment via the openspec CLI. These operations are scoped to project management tasks such as listing changes, checking status, and reading instructions.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it reads and parses external specification and task files (contextFiles).
  • Ingestion points: contextFiles (proposal, specs, design, tasks) and CLI JSON output.
  • Boundary markers: Absent.
  • Capability inventory: Bash(openspec:*) and file system modifications (implied for task implementation).
  • Sanitization: Absent. While a vulnerability surface exists, the potential impact is limited by the restricted command set and the intended use case of processing local project artifacts.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 18, 2026, 04:49 PM
Security Audit — agent-trust-hub — openspec-apply-change