event-credit-distribution
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources which could potentially contain malicious instructions or unexpected formatting.
- Ingestion points: The skill reads guest registration answers from Luma and response data from Google Sheets/Forms, specifically targeting participant-provided X handles.
- Boundary markers: Instructions require normalization of handles (trimming, case-insensitivity, and removing prefixes). The skill also uses a dedicated JavaScript helper (
render-credit-message.js) to process templates, providing a buffer between raw data and the final message. - Capability inventory: The skill possesses read capabilities for external APIs and spreadsheets, and write capabilities via XChat transports. It also executes local scripts via subprocesses.
- Sanitization: The
render-credit-message.jsscript performs validation to ensure placeholders like{{codex_url}}are resolved exactly once and that no nested URL prefixes or placeholder markers (e.g.,XXX) remain in the final output. - [DYNAMIC_EXECUTION]: The skill uses a local script to process data at runtime, which is a common but noteworthy execution pattern.
- Evidence: The instructions specifically direct the agent to use
node scripts/render-credit-message.jsto render the approved message template using JSON input from the manifest. This execution is scoped to a specific, bundled script with defined inputs and outputs.
Audit Metadata