harvest
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use shell commands via
npx @useclawlink/clito perform authentication, connection, and data operations within the Harvest integration. - [EXTERNAL_DOWNLOADS]: The skill requires the execution of
npx, which downloads the@useclawlink/clipackage from the npm registry (a well-known service) at runtime to perform its tasks. - [INDIRECT_PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection where untrusted data could influence agent behavior.
- Ingestion points: User-provided descriptions of actions are interpolated directly into shell command arguments in
SKILL.md(e.g.,npx @useclawlink/cli actions harvest "<what you want to do>"). - Boundary markers: There are no explicit delimiters or instructions to the agent to sanitize or ignore embedded control characters within the user-supplied strings.
- Capability inventory: The agent is granted the ability to execute shell commands and interact with the Harvest API (read/write access).
- Sanitization: The instructions do not specify any validation or sanitization steps for the input strings before they are passed to the shell.
Audit Metadata