dedot
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill runs a local Node.js script (
scripts/find.mjs) to locate U+00B7 characters within files or folders specified by the user. This operation is limited to reading file content for punctuation analysis. - [INDIRECT_PROMPT_INJECTION]: The skill reads and processes the content of external files, which are then presented to the agent for review. This creates a potential surface where instructions hidden in the analyzed files could influence the agent's behavior.
- Ingestion points: The script
scripts/find.mjsreads lines from user-specified paths and outputs them to the agent's context. - Boundary markers: There are no explicit delimiters or specific instructions to the agent to disregard commands found within the analyzed text segments.
- Capability inventory: The skill is scoped to punctuation review and generating text proposals. It is restricted from invoking other tools or models via the
disable-model-invocation: trueconfiguration in the frontmatter. - Sanitization: The skill does not perform sanitization or filtering of the text retrieved from files before it is processed by the agent.
Audit Metadata