API Risk Monitor
OpenAPI ingestion (JSON, YAML, HTTPS URL), risk score, structured findings, Markdown export, recent run history, and Stripe-backed access when billing is configured on the deployment.
Product
Named capabilities on this Next.js deployment: what is established in production, what shipped recently, what is planned next, and the ordered API Risk engineering backlog. Next cross-product cycle (MCP posture + mobile exploration) is summarized in docs/ROADMAP-NEXT.md. HTTP route detail for scans and compares lives in docs/API-RISK-FEATURES-AND-ROUTES.md.
Core product lines available on the live app today.
OpenAPI ingestion (JSON, YAML, HTTPS URL), risk score, structured findings, Markdown export, recent run history, and Stripe-backed access when billing is configured on the deployment.
Control-plane flows: policies, agent inventory, API keys, human-in-the-loop approvals, audit-oriented events, and dashboard entry points for operators.
Declarative manifest path, identity inventory surfaces, dashboard workflows, correlation hooks toward API posture, and tiered access patterns.
Capabilities that landed on the live app in the current release window.
Compare two OpenAPI documents (JSON, YAML, or URL per side) with score delta, route churn, auth drift, and findings appeared versus removed.
Compare runs stored alongside scans when a store file is configured; JSON detail route and Markdown report download for CI and auditors.
Response headers exposing score and finding counts on scan and compare success paths, plus a thresholded CI gate script for pipelines.
Snapshot history table, inventory CSV export, and optional workspace label on higher tiers for team-facing scope naming.
Planned product capability — scope is engineering-owned until shipped.
Next delivery focus: design-time MCP surface posture (tools/resources), policy templates, mapping toward OpenAPI-backed risk signals, stable IDs, and exportable reports; optional encryption for sensitive posture payloads. Written product plan in the Labs repo; cross-product notes in docs/ROADMAP-NEXT.md.
Exploration only — requirements, scope, and non-goals are not fixed yet; no implementation timeline. Documented intent so stakeholders know a client strategy will be decided deliberately after use-case review (see docs/ROADMAP-NEXT.md).
Ordered follow-ups after the shipped compare stack.
Tighten detection of example and default fragments that may embed secret-like strings inside published specifications.
Optional outbound notification when a compare job completes, aligned with existing scan automation patterns.
Optional retention TTL or per-plan cap on retained compare rows, separate from the scan list length limit.