Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+21 moreClaude CodeCursorCodex CLIMuseDots
Manus
Grok
Grok Bot+18 agentsMedical & Pharma AI Compliance Gate — MLR & Publication Read
Audit medical and pharma content for compliance-readiness, claim substantiation, and citation integrity.
$39.99
Medical & Pharma AI Compliance Gate — MLR & Publication Read
Example session with this skill installed
Review this draft HCP email for Vaxira (product) in the UK market. I've attached the SmPC. Check the claim 'Vaxira reduces recovery time by 40%' against Section 5.1 of the SmPC.
- Read your context and instructions
- Compiled the medical & pharma
I have completed the compliance-readiness audit for the Vaxira HCP email draft against the provided UK SmPC (dated Oct 2023).
Verdict: REVISE
Audit Report
| Finding ID | Location | Issue Type | Severity | Recommendation |
|---|---|---|---|---|
| FIND-001 | Paragraph 2 | Claim Substantiation | BLOCK | The 40% reduction claim exceeds the SmPC Section 5.1 data (32% in study A-102). Revert to SmPC wording. |
| FIND-002 | Paragraph 4 | Fair Balance | REVISE | Warning regarding hepatic impairment is missing from the footer. Add text from SmPC Section 4.4. |
| FIND-003 | Citation 1 | Reference Integrity | PASS WITH NOTES | DOI verified. Note: Study population in citation is limited to adults 18-65; ensure email doesn't imply geriatric use. |
| FIND-004 | Header | AI Disclosure | REVISE | Draft indicates AI-assisted drafting; ensure internal PharmaCo AI-use policy header is attached. |
Claim/Evidence Matrix
| Draft Claim | Source Location | Status | Evidence/Notes |
|---|---|---|---|
| "Vaxira reduces recovery time by 40%" | SmPC 5.1 | UNSUPPORTED | SmPC cites mean reduction of 32% (p<0.05). 40% figure appears to be a post-hoc subgroup analysis not in label. |
| "Well-tolerated in most patients" | SmPC 4.8 | REVISE | Vague terminology. SmPC lists "Common" (1/10) nausea. Suggest specific incidence rates for fair balance. |
Safety and Fair Balance Findings
The draft emphasizes efficacy in the hero section but places contraindications in a small-font footer. This visual imbalance may trigger MLR rejection under UK ABPI guidelines. Potential adverse event mentioned in patient testimonial requires immediate routing to Pharmacovigilance.
Next steps
- Correct the recovery time percentage to match the 32% figure in SmPC Section 5.1.
- Insert the mandatory Hepatic Impairment warning into the safety summary.
- Route the testimonial in Paragraph 5 to the PV department per SOP-04.
Connects securely to your tools. The creator never sees your data.
About this skill
Drafting medical and pharma content with AI introduces significant regulatory and clinical risks. This skill acts as a pre-submission compliance gate, auditing manuscripts and marketing materials before they reach formal Medical, Legal, and Regulatory (MLR) review. It identifies evidence gaps, safety imbalances, and citation errors that typically trigger costly revision cycles or regulatory rejections.
What it does
- Claim substantiation audits specific wording against approved labels, SmPCs, or evidence packs to flag off-label risks.
- Reference integrity verifies DOIs, PMIDs, and source consistency to detect misquotations or paraphrase drift.
- Fair balance check evaluates the presentation of contraindications and adverse reactions against benefit claims.
- AI disclosure review ensures human accountability and AI-usage statements align with ICMJE, GPP, or journal policies.
- Privacy screening flags direct identifiers or sensitive patient data that increase re-identification risks.
How it works
- Intake Provide the draft content, the intended audience, and the controlling source documents like an approved SmPC.
- Analysis The skill extracts claims and safety statements, cross-referencing them against your supplied evidence.
- Verdict It issues a PASS, REVISE, or BLOCK status based on the severity of findings.
- Handoff You receive a structured report including a claim-evidence matrix and a prioritized must-fix list.
Frameworks & tools
Built for healthcare communications, medical affairs, and regulatory teams using guidelines from ICMJE, GPP, and standard MLR workflows.
Why this beats prompting it yourself
Standard prompts often hallucinate medical evidence or miss subtle off-label nuances. This skill enforces a conservative, evidence-linked auditing logic that distinguishes between direct and indirect support while explicitly flagging unverified citations rather than guessing.
Use cases
- Auditing a pharmaceutical product website draft against the FDA-approved prescribing information.
- Checking a scientific manuscript's AI-disclosure and citation accuracy before journal submission.
- Reviewing internal medical education slides for fair balance and safety data inclusion.
- Screening AI-generated patient summaries for accidental PHI leakage.
Known limitations
This skill cannot perform external lookups for paywalled sources or make final legal/clinical decisions. It requires users to supply the controlling label or evidence pack for substantiation.
Requires user-supplied controlling sources (PI/SmPC). Cannot verify paywalled references or perform formal PV reporting. Statuses are readiness indicators, not legal approvals.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 24 days ago