delegate-imagegen
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the delegation of user-provided task descriptions to external worker agents. This creates a surface where instructions embedded in user data could potentially influence the worker's behavior.
- Ingestion points: User input provided via stdin to scripts such as
run-imagegen.shandprepare-imagegen.sh. - Boundary markers: The skill documentation advises agents to structure requests using Markdown headers (Objective, Scope, Constraints, etc.) to maintain task context.
- Capability inventory: The skill possesses the capability to execute shell commands via the
Bashtool to run local scripts and a Node.js CLI. - Sanitization: Sanitization logic is expected to be handled by the internal
delegate-cli.mjscomponent; no explicit filtering is present in the shell wrapper scripts.- [EXTERNAL_DOWNLOADS]: The skill references official pricing and documentation URLs for major AI providers including Anthropic, OpenAI, Cognition, and Cursor. These references are used for informational cost analysis and do not involve the execution of remote code or the download of untrusted assets.
Audit Metadata