$ loading_
Audit support reply drafts for missing policy disclosures and suggest fixes.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Behavioral Audit MCP" yet — see the docs or source repo.
Please audit this support reply draft and determine whether it misses any refund or policy disclosures relevant to the customer's situation. Return a pass/gap verdict and suggested revisions: Customer situation: The user was charged after the trial ended and is requesting a refund. Draft reply: We're sorry for the inconvenience, and we can help process your refund request.
A verdict on whether the reply passes, plus missing disclosures and a more complete suggested reply.
Review the following support reply draft for any missing policy disclosures relevant to an account suspension appeal. Output a pass/gap verdict and list suggested additions: Customer situation: The user's account was suspended for unusual activity, and they submitted an appeal. Draft reply: We have received your appeal and will review it as soon as possible before getting back to you.
A verdict showing whether disclosure gaps exist, along with what policy information should be added.
Audit this customer support template for completeness of policy disclosures. Determine whether anything is missing for the specific customer scenario and suggest improvements: Customer situation: The customer requests account and related data deletion. Draft reply: We have received your deletion request and will arrange processing for you.
A pass/gap result, plus policy disclosure suggestions better suited to the scenario.
Support or operations staff can review draft replies before sending, especially for refunds, appeals, or account-related requests. This helps catch policy disclosures that are relevant to the customer's situation but missing from the draft.
Team leads can use it to identify disclosure gaps in draft replies and standardize suggested improvements. It is useful for improving consistency and completeness in customer support communication.
It checks draft support replies for policy disclosures that may be missing based on the customer's specific situation. It returns a pass or gap verdict along with suggestions.
Based on the description, it needs at least a draft support reply and information about the customer's situation. It uses that context to determine which policy disclosures are relevant.
It returns a pass/gap verdict together with suggestions for improvement. For the exact output format, see the source repository.
Audit Microsoft 365 privacy, compliance, and security settings with natural language.
Audit liability clauses against company standards with evidence-based verdicts or abstention.
Audit LLM-as-judge results for drift, bias, and human agreement.
Audit app store metadata and refine keyword fields inside MCP clients.
Handle support tasks like order lookup, refunds, and audit review safely.
Audit MCP configs for exposed access, secrets, models, and compliance AI-BOMs.