convex-expert
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to execute shell commands for code validation and deployment, specifically
npx tsc --noEmitandnpx convex dev --once. These are standard tools for projects using the Convex platform. - [INDIRECT_PROMPT_INJECTION]: The skill is configured to read and process data from local developer-controlled files, which could theoretically contain instructions designed to influence the agent's behavior.
- Ingestion points: The skill reads
convex/schema.tsandconvex/_generated/ai/guidelines.mdto establish context. - Boundary markers: There are no explicit instructions or delimiters provided to the agent to treat the content of these files as untrusted data or to ignore embedded instructions.
- Capability inventory: The agent has the capability to execute shell commands (
npx) and perform network operations via the Convex CLI as part of its deployment workflow. - Sanitization: The skill does not specify any sanitization or filtering logic for the content ingested from the local project files.
Audit Metadata