openspec-sync-specs
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to "intelligently merge" changes by reading delta specifications and directly editing main specification files. Because it interprets natural language requirements and scenarios from these files, there is a risk that untrusted or malicious content within the specs could contain instructions intended to influence the agent's behavior during the editing process.
- Ingestion points: Reads delta specification files containing requirement and scenario descriptions (SKILL.md, Step 3 & 4).
- Boundary markers: The skill does not define specific delimiters or warnings to ignore instructions embedded within the specification content.
- Capability inventory: The skill has the ability to read and write files within the
openspec/specs/directory and execute CLI commands. - Sanitization: No explicit sanitization or validation of the specification content is mentioned before it is processed by the agent.
- [COMMAND_EXECUTION]: The skill utilizes the
openspeccommand-line interface to list changes and retrieve status information (openspec list --json,openspec status --change "<name>" --json). These commands are essential for resolving the context of the specifications being synchronized.
Audit Metadata