openspec-continue
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements a workflow where the agent reads existing artifact files (like proposal.md or design.md) to generate the next items in a dependency chain. This creates a surface where instructions embedded in a processed file could potentially influence the agent's behavior during the creation of subsequent files.
- Ingestion points: The agent reads local dependency files for context as described in the 'Workflow' and 'Create one artifact' sections of SKILL.md.
- Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following natural language instructions found within the ingested artifacts.
- Capability inventory: The skill utilizes the
openspecCLI for status checks and instruction retrieval, and it has the capability to create and write new files. - Sanitization: The workflow does not include steps for sanitizing or validating the content of the artifacts before they are processed by the agent.
Audit Metadata