// MCP TOOL REFERENCE

MVR API - MINIMUM VIABLE RELATIONSHIPS TOOLS

19 tools from the MVR API - Minimum Viable Relationships MCP Server, categorised by risk level.

View the MVR API - Minimum Viable Relationships policy → Token cost: 2,216 tokens →

// ALL 19 MVR API - MINIMUM VIABLE RELATIONSHIPS TOOLS

READ 13 tools

Read mvr_abstention_check Abstention guardrail. Decide whether to refuse, soften, or caveat a confident recommendation when relationa... Read mvr_accelerator_screen Accelerator screening preflight. Separate claimed traction from evidence-ready applicants before cohort sel... Read mvr_auth_check Validate an MVR key or the public demo key. Use only when an agent needs to confirm access mode before call... Read mvr_decision_check Produce the final MVR readiness verdict, from permission_not_yet_earned to ready_to_scale. Use last, once e... Read mvr_evidence_completeness Check if there is enough evidence to decide yet and list missing proof. Use before decision_check when evid... Read mvr_get_benchmark_scenarios Find public MVR-Bench dev scenarios and private-test boundaries. Use for benchmark setup, not for live mark... Read mvr_get_leaderboard Fetch public MVR-Bench leaderboard rows, scores, and Reckless-GO Rate baselines for citation or benchmark r... Read mvr_get_reckless_go_rate Explain Reckless-GO Rate: how often an agent greenlights readiness beyond evidence. Use when the user asks ... Read mvr_guardian_risk_scan Guardian-risk scan for regulators, gatekeepers, local authorities, community custodians, or veto holders th... Read mvr_investor_due_diligence Investor memo preflight. Flags missing relational proof before funding, diligence, valuation, or investment... Read mvr_market_permission_memo Market-permission memo builder. Summarize safe claims, unsafe claims, blocking dimensions, and next proof f... Read mvr_partner_readiness Local-partner readiness check. Use when the question is whether a distributor, channel, implementer, or par... Read mvr_preflight_market_entry Market-entry guardrail for launch or expansion advice. Use before recommending entry into an African or hig...

WRITE 3 tools

Write mvr_entity_resolve Canonicalize the venture, market actor, country, sector, and archetype. Use first when the user gives a nam... Write mvr_evidence_gap_plan Evidence gap planner. Return the next proof to request before the agent makes a stronger claim or go/no-go ... Write mvr_submit_benchmark_run Prepare an MVR-Bench run submission. Public MCP returns instructions; private leaderboard scoring stays ser...

DESTRUCTIVE 1 tools

Destructive mvr_agent_market_entry_guardrail Drop-in agent guardrail before market entry, MVP launch, scale, funding, CFO action, or partnership advice ...

EXECUTE 2 tools

Execute mvr_context_compile Turn formal, informal, sentiment, field, and market context into safe and unsafe inferences. Use after evid... Execute mvr_preflight_mvp MVP-advice guardrail. Use before telling a founder to build when trust, permission, embeddedness, or guardi...

// FAQ

How many tools does the MVR API - Minimum Viable Relationships MCP server have? +

The MVR API - Minimum Viable Relationships MCP server exposes 19 tools across 4 categories: Read, Write, Destructive, Execute.

How do I enforce policies on MVR API - Minimum Viable Relationships tools? +

Route the MVR API - Minimum Viable Relationships server through the PolicyLayer gateway. Define allow, deny, or approval rules per tool in the dashboard — they are enforced on every call before it reaches the server.

What risk categories do MVR API - Minimum Viable Relationships tools fall into? +

MVR API - Minimum Viable Relationships tools are categorised as Read (13), Write (3), Destructive (1), Execute (2). Each category has a recommended default policy.

Let agents act without letting them run wild.

Route your MCP servers through PolicyLayer and every tool call is checked against your policy before it runs — allow, deny, or require approval. Per-identity grants. Full audit log. Live in minutes.

SECURE YOUR MCP →

Free to start. No card required.

4,600+ MCP servers and 31,000+ tools scanned and risk-classified.