openspec-propose
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
openspecCLI tool to create new change directories and retrieve status and instruction data. - [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it processes instructions and rules retrieved from an external CLI tool to guide its behavior.
- Ingestion points: User-provided change descriptions and JSON-formatted data returned by the
openspec instructionscommand. - Boundary markers: The agent is explicitly instructed to treat
contextandrulesas internal constraints rather than content to be copied into the final artifacts. - Capability inventory: The skill has the ability to execute local shell commands via the
openspecCLI and create files on the local file system. - Sanitization: The agent is instructed to sanitize user input into a kebab-case format and follow predefined templates when writing to the output paths.
Audit Metadata