incipient.ai
← Intelligence Hub

Governance · 2026-06-16 · 11 min

Governing MCPs and Agents — Metadata, Lifecycle, and Risk-Tiered Approval

A common metadata model, one registration-to-retirement state machine, a risk score computed from sensitivity × autonomy × action × reach, and approval paths that scale review depth to risk.

Governing the agentic layer comes down to three mechanisms: describe every asset the same way, move every asset through one lifecycle, and scale review depth to computed risk. Get those right and an agent marketplace is governable at scale; get them wrong and it is a catalog of unaccountable software.

Describe everything — the metadata model

Every asset shares a common metadata core, extended by type. Nothing gets published without it.

Common core — all assets:

  • Asset ID, name, type (MCP / agent / agent team)
  • Purpose — plain-language what & why
  • Business area(s) + journey & stage mapping
  • Business owner + technical owner (both named)
  • Owning org / cost center (for chargeback)
  • Status, version & changelog
  • Certification tier (Sandbox → Gold) & risk tier
  • Data classification touched
  • SLA, support channel, cost per invocation
  • Recertification due date

MCP-specific adds the connection and safety surface: endpoint, transport & environment; auto-introspected tool manifest; auth method & identity propagation; systems & data products connected; least-privilege scopes with action tools flagged; rate limits & quotas per consumer; prompt-injection / output-safety posture.

Agent-specific adds the autonomy and evaluation surface: agent pattern & autonomy level (L1–L4); model dependencies (allowlist); MCPs & tools consumed (dependency graph); human oversight points; versioned prompt/instruction register; eval results & red-team evidence; memory & data retention; escalation, fallback & regulatory scope.

The "cost per invocation" and "owning cost center" fields matter more than they look — they are what make the marketplace's economics real rather than notional.

Move everything through one lifecycle

MCPs and agents share a single state machine; the risk tier only changes how deep the review is, not the states:

Draft → Submitted → In review → Approved → Published → Deprecated → Retired
StateWhat happens
DraftIntake form or CI/CD registration; metadata capture
SubmittedDuplication check; manifest introspection
In reviewPolicy-as-code gates + risk-tiered human review
ApprovedConditions possible: autonomy caps, pilot scope
PublishedSearchable, badged; access via entitlements
DeprecatedSunset announced; consumers get a migration path
RetiredAccess revoked; immutable audit record kept

Two rules keep it honest: any new tool, scope expansion, model swap, or autonomy increase re-enters review (a lighter delta review for low/medium risk), and a guardrail or security incident can auto-suspend a published asset. Automated validation runs at submission — model allowlist, auth pattern, secrets vaulted, classification-vs-scopes consistency, vulnerability scan, action-tool inventory.

Compute risk, then scale the review

Risk is not a gut call. It is computed:

risk = data_sensitivity × autonomy_level × action_capability × user_reach
     → Low / Medium / High / Critical

An agent that only suggests over public data is Low; an autonomous agent transacting on regulated systems with broad reach is Critical. The tier sets the approval path and SLA:

Risk tierReview pathSLAExample
LowDomain steward (single approver)2 business daysRead-only MCP over public product data
MediumDomain steward + Security review5 business daysAgent drafting content, human-reviewed output
HighAI Governance Board (Security, Privacy, Legal, Domain, Architecture)10 business daysPHI-adjacent audience agent with action tools
CriticalGovernance Board + executive sign-off + mandatory red-team eval15–20 business daysAutonomous agent transacting on regulated systems

Reviewers all see the same review dossier: metadata, auto-check results, eval scores, dependency graph, and a similar-asset comparison. Outcomes are bounded: approve / approve with conditions / return with findings / reject.

Who does what — clear accountability

Every stage has an owner, on a simple RACI:

RoleResponsibility
Requestor / builderRegisters the asset, supplies metadata and eval evidence, responds to findings, attests at recertification
Business ownerAccountable for business fit, purpose, and value; signs off publication, recertification, retirement
Domain stewardApproves low-risk assets; curates the domain's journey catalog; first line for duplication and quality
Security & privacyReviews medium+ risk: classification, PII/PHI handling, auth, prompt-injection posture
AI governance boardApproves high & critical risk; owns policy-as-code rules, model allowlist, certification standards
Platform teamRuns the marketplace: automated validation, telemetry, dependency graph, access provisioning, retirement mechanics

Recertification keeps it from rotting: owners attest quarterly (Gold) to annually (Bronze), and a lapsed attestation auto-deprecates the asset.

The path to implement

  1. Make complete metadata mandatory at intake — no metadata, no publication.
  2. Implement the single lifecycle state machine with automated validation at submission.
  3. Compute the risk tier from sensitivity × autonomy × action × reach.
  4. Route to risk-tiered approval paths with explicit SLAs and one shared review dossier.
  5. Enforce recertification so lapsed attestation auto-deprecates.

This is human-in-the-loop governance and policy-as-code compliance applied to the agentic layer. Incipient's GovPilot and Data Marketplace accelerators provide the policy engine, audit trail, and registry to run it.

Talk to us about agent governance →