Open findings, grouped by the policy producing them. Every finding is independently addressable; grouping never merges them. A finding is Enforced when Qpoint blocked it, and then it is resolved in the same moment as prevented; it is Observed when it was only recorded.

Severity		Findings	Harnesses	Relevant Frameworks
critical	Credentials must not be sent to a model	1 High 1 past SLA	Cursor CLI	OWASP LLMOWASP AgenticNIST AI RMFSOC 2
high	Sensitive information sanitized before transmission	3 High 3 past SLA	Cursor CLIClaude Code	OWASP LLMOWASP AgenticNIST AI RMFISO 27001SOC 2
high	Personal account is in use	1 High	Claude Code	OWASP AgenticNIST AI RMFSOC 2
medium	Prompts are inspected for injection and leakage	1 Med	Claude Code	OWASP LLMOWASP AgenticNIST AI RMF
medium	Consequential tool calls require approval	1 Med	Claude Code	OWASP LLMOWASP AgenticNIST AI RMFSOC 2

3 policies have no findings under these filters and are not listed.