openspec-sync
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the
specs/directory of an active change and merges it into theopenspec/specs/main directory. This represents a surface for indirect prompt injection if the specifications contain malicious instructions intended to influence the agent during subsequent processing steps. - Ingestion points: Specification files within the active change's
specs/folder as described in the workflow. - Boundary markers: No specific delimiters or safety warnings for the spec content are defined in the instructions.
- Capability inventory: The skill performs file-write operations to the main
openspec/specs/directory. - Sanitization: There is no evidence of content validation or escaping of the ingested specification data.
- [EXTERNAL_DOWNLOADS]: The skill references project documentation located on an external platform.
- Source: Technical documentation hosted at
github.com/Fission-AI/OpenSpec.
Audit Metadata