# SCRIMED Product Readiness Brief

Status: commercial-pilot-ready
Boundary: SCRIMED is sellable today as a governed synthetic pilot and enterprise operating-system evaluation surface; live clinical execution remains gated until identity, runtime safety, durable audit, privacy, connector, and human-review controls are approved.

## Enterprise Offers
- Synthetic Pilot Evaluation: A 30 to 90 day synthetic pilot with workflow maps, agent scopes, validation evidence, risk controls, and executive findings.
- Workflow Intelligence Assessment: Workflow inventory, friction map, automation candidate scorecard, interoperability targets, and human-review design.
- AI Readiness + Governance Audit: Governance gap report covering privacy posture, auditability, role controls, runtime safety, model/workflow oversight, and approval gates.
- Clinical Operations Automation Blueprint: Prioritized automation roadmap with agent responsibilities, connector plan, review queues, safety boundaries, and phased deployment path.

## Pricing and Sales
Pricing route: /pricing
Pricing API: /api/commercial/pricing
Recommended model: Hybrid enterprise model: free public preview, paid assessment, paid synthetic pilot, protected enterprise pilot, annual platform license, and custom strategic partnerships.
- Public Product Preview: Free public access
- Workflow Intelligence Assessment: Recommended range: $15k-$35k fixed fee
- Synthetic Pilot Evaluation: Recommended range: $35k-$95k for 30-90 days
- Protected Enterprise Pilot: Recommended range: $150k-$400k for 90-180 days
- Enterprise Operating License: Annual license: custom, recommended floor $300k+
- Strategic Platform Partnership: Custom multi-year partnership

## Operations Readiness
Operations route: /operations
Operations API: /api/operations/readiness
Status: blocked-actions-visible
Blocked actions: 1
Manual actions: 2
- GitHub push authentication is configured for the current workspace.: Re-run GitHub device auth or use the connected GitHub app for a connector-backed branch after accepting the history reconciliation plan.
- Local npm, pnpm, yarn, corepack, and tsc are unavailable in the current Codex PATH.: Continue using `git diff --check`, typed App Router patterns, and Vercel remote builds as the managed quality path.
- Vercel Git deployment path is working from GitHub `main`.: Use the Vercel dashboard or connector inspection if the CLI is unavailable.
- `app.scrimedsolutions.com` is recommended but not yet verified as connected to the Vercel app.: Use the current Vercel deployment URL for internal review until branded DNS is ready.
- Wix CTAs need to be wired to SCRIMED product routes.: Share direct Vercel links manually during buyer conversations.
- Public vs protected route policy needs an explicit decision before broad buyer access.: Use Vercel share links or authenticated demos for early enterprise review.

## Agents
- Prior Authorization Agent (planned): Prepare reviewable prior authorization packets from structured clinical, coverage, and policy context.
- Revenue Cycle Agent (planned): Identify documentation, coding, claim, and denial-workflow gaps without making final billing determinations.
- Scheduling Agent (planned): Coordinate scheduling options across clinical urgency, patient preference, eligibility, and operational capacity.
- Trial Matching Agent (staged): Map synthetic patient signals to trial eligibility questions and produce reviewable matching rationale.
- Documentation Agent (staged): Produce reviewable clinical documentation drafts from structured context while preserving clinician authorship.
- Compliance Agent (foundation): Check workflows against privacy, security, auditability, and human-review requirements before expansion.
- Interoperability Agent (foundation): Map workflow needs to FHIR, HL7, claims, pricing, device, and future healthcare connector contracts.
- Clinical Intelligence Agent (planned): Summarize clinical context and surface reviewable decision-support signals without replacing clinical judgment.
- Research Agent (planned): Synthesize research context, evidence gaps, and cohort questions for human research teams.
- Governance Agent (foundation): Track model, workflow, policy, risk, and quality-gate changes across SCRIMED operating surfaces.
- Supply Chain Agent (planned): Detect supply, inventory, vendor, and care-continuity risks before they affect patient operations.

## Workflow Engine
- Referral intake automation: Referral workqueue state, missing-evidence list, routing rationale, and escalation reason.
- Prior authorization support: Packet draft, cited policy rationale, missing-evidence list, and reviewer approval state.
- Patient onboarding triage: Navigation recommendation, urgency rationale, Watchtower trace, and human-review requirement.
- Ambient documentation review: Draft note, source trace, missing-data prompts, review checklist, and blocked final-signature state.
- RCM denial risk review: Denial-risk rationale, documentation gap list, appeal draft outline, and coding-review queue.
- Care gap detection: Care-gap signal list, context summary, review prompt, source trace, and escalation boundary.

## Governance Controls
- Human review required: Every staged workflow and agent action remains review-gated before external, clinical, payer, or patient-facing use.
- Synthetic data only: Current pilots use deterministic synthetic fixtures and do not ingest production clinical records.
- No autonomous diagnosis: SCRIMED surfaces operational intelligence and review prompts; it does not diagnose or replace clinician judgment.
- Audit trail enabled: Workflow, result, denial, and quality surfaces retain inspectable traces and metadata-only evidence boundaries.
- HIPAA-ready posture: Architecture is designed toward HIPAA-grade privacy, security, and audit controls before live protected health information.
- Privacy-by-design: The product keeps clinical execution gated, avoids request-body capture in denied execution paths, and minimizes data exposure.
- Role-based access planned: Production workflow use requires approved tenant identity, role permissions, service auth, consent, and break-glass policy.

## AgentOS v1
Status: synthetic-agent-platform-ready
Control plane agents: 4
Task engine: /api/agent-os/tasks
Evaluation workspace: /evaluation
Evaluation API: /api/agent-os/evaluation

## Atlas Intelligence Core v1
Status: continuous-validation-operating-layer
Subsystems: 7
Trust Cards: 3

## Evidence Metrics
- Time saved: Time savings must be measured in buyer pilots against approved baseline workflows.
- Workflow friction reduced: Friction reduction is an operational pilot metric, not a live-care outcome claim.
- Documentation quality improved: Quality improvement requires clinician review and buyer-approved documentation scoring.
- Revenue leakage identified: Financial impact must be validated against buyer revenue-cycle data under approved controls.
- Patient access bottlenecks surfaced: Access impact is measured during protected pilots after privacy, consent, and review workflows are approved.

## Next Commercial Move
Use the AgentOS Evaluation Workspace to turn synthetic buyer packets into Atlas Trust Cards, task plans, audit previews, observability records, and a production-readiness decision register.

## Enterprise Intake
Pilot intake route: /pilot
Pilot intake API: /api/pilot/intake

Updated: 2026-06-02