sales-qwilr-automation
Pass
Audited by Gen Agent Trust Hub on Apr 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends installing an additional tool using
npx skills add sales-skills/sales. This resource originates from the same vendor ('sales-skills') as the skill itself. - [DATA_EXFILTRATION]: The skill processes sensitive information including CRM data and Qwilr API tokens. It mitigates credential exposure risks by explicitly instructing the user to use environment variables (
$QWILR_TOKEN) rather than hardcoding secrets. It also logs operational knowledge toreferences/learnings.md. - [PROMPT_INJECTION]: The skill is subject to indirect prompt injection because it ingests data from external sources (CRMs and webhooks) to populate document templates.
- Ingestion points: External CRM fields (e.g., Salesforce, HubSpot) are mapped to Qwilr tokens, and webhook payloads are processed for automation triggers as described in
SKILL.mdandreferences/qwilr-api-reference.md. - Boundary markers: The skill does not define specific delimiters or instructions to ignore potential malicious content embedded within the CRM data before it is interpolated into the
{{token}}placeholders. - Capability inventory: The skill is designed to perform network requests to the Qwilr API (
api.qwilr.com) and perform local file writes toreferences/learnings.md. - Sanitization: There is no evidence of data sanitization or validation logic to filter potentially malicious instructions from the CRM or webhook inputs.
Audit Metadata