Skip to main content
Glama
nonameuserd

@chitmark/haven-mcp

by nonameuserd
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Hybrid

list_capabilities

Read-onlyIdempotent

Fetch read-only capability catalog to rank host tools and Haven agent delegation before session creation. Apply routing policies (best, as_provided, constrained_best) and evidence gates for auditable decisions.

Instructions

Read-only machine-readable capability catalog Haven publishes for host merge. Returns haven.agent_delegation plus hostMerge.guide (scoreHints cookbook + examples) and an auditable routing ranking. Policies: best (soft weighted), as_provided (caller order), constrained_best (hard constraints then lexicographic objective; requires constraints + optional objective; emits ranking.filtered). Structured evidence gates (verification.status, scope.domain, freshness, verifierTrust) use component cards; never evidenceConfidence. Soft peerWarnings when host peers omit hints. Median completion latency is never ranked (fact + measuredN only). Optional task improves fit. Optional peers ranks host tools beside Haven (POST /api/capabilities/rank). Never forces Haven selection, never means fail-over after vendor failure, and never shuffles. Call before create_session when deciding whether agent-delegation fits.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
asOfNoISO timestamp for freshness age/expiry evaluation (deterministic). Default: now.
taskNoOptional task text used for fit scoring (e.g. what you need done).
peersNoOptional host peer capability cards to rank beside Haven. Same layout as catalog entries; forceSelection must be false. Attach structured evidence cards (capability/claim/verification/freshness/scope).
policyNoRouting policy. best (default): soft weighted rank. as_provided: preserve caller order. constrained_best: hard constraints then soft objective (requires constraints). random/shuffle are rejected.
objectiveNoSoft lexicographic objective among feasible candidates. Example: { "maximize": "fit", "secondary": "minimize expectedSteps" }.
constraintsNoHard eligibility gates for policy=constrained_best. Score dims (">= 0.70", "== compatible") plus evidence components (verification.status, scope.domain, freshness.ageDays, verifierTrust, evidenceProvenance). No evidenceConfidence. Infeasible candidates appear in ranking.filtered.
includeHavenNoWhen peers are supplied, include Haven's agent_delegation card (default true).
minProvenanceNoFirst-class provenance floor for policy=constrained_best (compiles to evidenceProvenance >= level; eliminations audit as provenance_below_min). Rejected on other policies.
trustedVerifiersNoHost trust list for verifierTrust: "== trusted". Required when that constraint is set. Unknown/adversarial verifiers fail closed.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changedv0.1.20
    • addedInput schema / properties / minProvenance
      Added value: +{
+  "description": "First-class provenance floor for policy=constrained_best (compiles to evidenceProvenance >= level; eliminations audit as provenance_below_min). Rejected on other policies.",
+  "enum": [
+    "self_attested",
+    "observed_attributable",
+    "independently_verified"
+  ],
+  "type": "string"
+}
  2. Addedv0.1.17

TDQS

A4.6/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already cover readOnly and idempotent, so the safety profile is handled. The description adds substantial behavior beyond those: policy semantics with ranking.filtered, evidence gates that never use evidenceConfidence, latency never ranked, soft peerWarnings, and strong guarantees like 'never forces Haven selection' and 'never shuffles.' This is far more than the annotations alone provide.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is dense but front-loaded with the core purpose, and every sentence contributes behavioral, usage, or policy detail. It is long, but each sentence earns its place; there is no filler or redundancy.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

There is no output schema, so the description compensates by naming return contents: haven.agent_delegation, hostMerge.guide, and a routing ranking, plus ranking.filtered for constrained_best. It covers policies, evidence components, peer warnings, and sequencing, making the tool self-sufficient for correct invocation and interpretation.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The schema covers 100% of the parameters, so the baseline is 3. The description adds meaningful context on how parameters interact, such as 'Optional task improves fit', 'Optional peers ranks host tools beside Haven', and constrained_best requiring constraints and emitting ranking.filtered. These augment the schema without simply repeating it.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific verb and resource: it is a read-only machine-readable capability catalog for host merge, returning haven.agent_delegation plus hostMerge.guide and a routing ranking. It clearly distinguishes this tool from the session-oriented siblings by instructing to call it before create_session when deciding whether agent-delegation fits.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It explicitly says 'Call before create_session when deciding whether agent-delegation fits,' giving a clear trigger condition. It also clarifies when constrained_best applies and rejects random/shuffle, and it warns that 'never means fail-over after vendor failure' and 'never shuffles.' It does not enumerate exclusions for every sibling, but the pre-session context is explicit.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools