openspec-new-change

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on the openspec CLI to manage the project lifecycle. It invokes commands such as openspec new change, openspec status, and openspec instructions to interact with the project structure and metadata. These operations are restricted to the local development environment and the openspec/ directory.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user descriptions to define a 'change name' which is subsequently used in shell commands. The instructions include a specific sanitization rule to derive a 'kebab-case' name, which effectively filters special characters and reduces the risk of command injection or file system traversal.
    • Ingestion points: User input for naming or describing the change requested in Step 2 of SKILL.md.
    • Boundary markers: Not explicitly defined for the shell command interpolation.
    • Capability inventory: Shell command execution (openspec CLI) and file system scaffolding within the project directory.
    • Sanitization: Explicit instruction to derive a 'kebab-case' name (e.g., add-user-auth) before proceeding to the command execution step.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 08:20 PM
Security Audit — agent-trust-hub — openspec-new-change