brida-reflex-sales-lead-fit

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill is instructed to fetch live availability and configuration data from the vendor's infrastructure using files such as /reflex/llms.txt, /reflex/agent.md, and /reflex/agent.json.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external lead data (facts) which is potentially untrusted content.
  • Ingestion points: Lead facts and firmographic data are ingested from the project environment, conversation, or connected tools as defined in the SKILL.md and references/playbook.md files.
  • Boundary markers: The skill uses a structured JSON contract (references/custom-reflex.json) to define the exact versioned public contract and expected output branches.
  • Capability inventory: The skill instructions explicitly state that the Reflex result never grants permission for protected actions such as payments, deploys, external messages, or account changes.
  • Sanitization: The instructions require that parsing, policy, and authorization remain in deterministic host code, and the data class is restricted to non-sensitive information.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 08:21 PM
Security Audit — agent-trust-hub — brida-reflex-sales-lead-fit