pursuit-brief
Pass
Audited by Gen Agent Trust Hub on Feb 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through external data processing.
- Ingestion points: The
generateAISectionsfunction inSKILL.mdinterpolates untrusted text fromrfp.titleandrfp.descriptiondirectly into the prompt string sent to the AI provider. - Boundary markers: There are no delimiters (e.g., XML tags or triple quotes) or explicit instructions to the AI to ignore embedded commands within the RFP data.
- Capability inventory: The skill has the ability to read from the internal RFP and evaluation databases and write generated briefs back to the
pursuitsdatabase via Convex mutations. - Sanitization: No evidence of input sanitization or validation exists to filter potential malicious instructions embedded in RFP documents.
Audit Metadata