tinacms
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONMETADATA_POISONING
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes Markdown and MDX content, creating a potential attack surface for indirect prompt injection.
- Ingestion points: Data is read from content directories like content/posts/ and content/docs/ into the agent context.
- Boundary markers: No delimiters or instructions are used to prevent the agent from interpreting content as instructions.
- Capability inventory: The skill is granted Bash, Write, and Edit tools.
- Sanitization: No input validation or sanitization is implemented for managed data.
- [DYNAMIC_EXECUTION]: The helper script scripts/check-versions.sh uses node -e for metadata parsing from package.json. While legitimate, this follows a dynamic execution pattern.
- [METADATA_POISONING]: An inconsistency exists between the reported author handle 'ma1orek' and the metadata 'Jeremy Dawes' in the plugin configuration.
Audit Metadata