openspec-sync-specs

Pass

Audited by Gen Agent Trust Hub on Oct 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • Command Execution via External Tool: The skill utilizes npx openspec to list and manage specifications. This involves running an external command-line utility, which is a standard pattern for this type of automation but relies on the security and integrity of the external package being executed.
  • Indirect Prompt Injection Surface: The skill is designed to read "delta specs" from markdown files and interpret their contents to update "main specs." Because the agent treats the text within these files as instructions for how to modify other files, there is a potential for indirect prompt injection.
  • Ingestion points: Specification files located at openspec/changes/<name>/specs/*/spec.md are read into the agent's context.
  • Boundary markers: The instructions do not define specific delimiters or "ignore instructions" markers to separate specification data from agent instructions.
  • Capability inventory: The agent has the ability to execute shell commands (npx) and perform file system read/write operations.
  • Sanitization: There are no explicit sanitization or validation steps mentioned for the content read from the delta specs before it is processed.
  • File System Interaction: The skill performs read and write operations on specific project directories. While necessary for synchronizing content, it means the agent modifies project documentation based on its interpretation of user-provided or file-based input.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 12, 2026, 09:55 PM