openspec-explore
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the project's OpenSpec CLI tool by executing the
openspec list --jsoncommand to gather information about the current project status and active changes. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from the local codebase and OpenSpec artifacts, which presents a surface for indirect prompt injection.
- Ingestion points: The agent is instructed in SKILL.md to read content from
openspec/changes/<name>/proposal.md,design.md, andtasks.md, as well as general codebase files. - Boundary markers: No explicit delimiters or instructions to ignore instructions embedded in those files are provided.
- Capability inventory: The skill allows for local CLI command execution (
openspec list --json) and broad file system read and search access as described in SKILL.md. - Sanitization: There is no evidence of sanitization or validation of the data retrieved from external files.
Audit Metadata