openspec-explore

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local codebase, configuration files (e.g., openspec/config.yaml), and instructions generated by the openspec CLI tool.
  • Ingestion points: Reads project source files, repository configuration, and command outputs from openspec status and openspec instructions.
  • Boundary markers: The instructions do not specify the use of delimiters or ignore-markers when reading external file content or tool outputs.
  • Capability inventory: The agent can read codebase files and write documentation artifacts (proposals, designs, tasks) via the openspec CLI.
  • Sanitization: No explicit sanitization or validation of content read from the file system or tool outputs is performed before it is processed by the agent.
  • [COMMAND_EXECUTION]: The skill uses the openspec CLI to perform actions like listing registered stores, initializing new changes, and retrieving artifact instructions. These commands are restricted to the openspec binary via the allowed-tools configuration.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 12:16 PM
Security Audit — agent-trust-hub — openspec-explore