finta-core-workflow-b

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for ingesting data from external financial and equity management platforms (Stripe, Mercury, Carta, Pulley) as well as CSV files.
  • Ingestion points: Workflow steps in SKILL.md involve importing metrics and cap table data.
  • Boundary markers: The instructions explicitly mandate the preparation of redacted drafts and require manual verification of recipient authorization before any data is sent.
  • Capability inventory: The skill is scoped to standard file manipulation tools (Read, Write, Edit, Grep).
  • Sanitization: The workflow includes human-in-the-loop checkpoints, requiring approval from an internal owner before finalizing updates.
  • [SAFE]: The skill promotes security best practices by instructing the agent not to store sensitive information like cap-table exports or recipient-level details in local repository files, instead keeping them within authorized production systems.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:44 AM
Security Audit — agent-trust-hub — finta-core-workflow-b