devin-handoff
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a standard integration for a well-known service (Devin). It correctly uses environment variables for API key management and communicates exclusively with official endpoints (devin.ai).
- [COMMAND_EXECUTION]: The skill executes local commands such as git, curl, and jq. These operations are used for context gathering and API interaction. The script safely constructs JSON payloads using jq, which prevents injection vulnerabilities when handling potentially complex code diffs.
- [INDIRECT_PROMPT_INJECTION]: The skill handles untrusted data from the local repository (specifically code diffs). It mitigates injection risks by encapsulating this data within structural delimiters like XML tags and markdown code blocks before sending it to the remote AI agent.
Audit Metadata