openspec-sync-specs
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
openspecCLI (e.g.,openspec list,openspec status) to manage specification changes. These operations are local and use parameters sourced from user input or CLI status reports. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by reading and processing external "delta spec" files.
- Ingestion points: Data enters the context during the reading of delta spec files in Step 4a of SKILL.md.
- Boundary markers: None identified; file content is processed directly without protective delimiters.
- Capability inventory: The agent can perform filesystem writes to the
openspec/specs/directory and execute local CLI commands. - Sanitization: No validation or escaping is applied to the content extracted from the delta specs before it is written to main specification files.
Audit Metadata