openspec-apply-change

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is exposed to indirect prompt injection because it processes untrusted content from the local project environment and the openspec CLI.
  • Ingestion points: The agent is instructed to read all files listed in the contextFiles array provided by the CLI, as well as the context and operationGuidance fields from JSON responses.
  • Boundary markers: The instructions do not define specific delimiters or "ignore previous instructions" markers when reading external file content, which could allow embedded directives to influence the agent.
  • Capability inventory: The skill has the capability to execute shell commands via the openspec CLI and modify source code and task files.
  • Sanitization: The skill contains protective logic that requires the agent to prioritize "built-in instructions" and "controlling values" from the CLI over external context or guidance if a conflict occurs, providing a layer of defense against malicious overrides.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 10:25 AM
Security Audit — agent-trust-hub — openspec-apply-change