Skip to main content
Glama

biii

by io.github.philpof102-svg

till_trust

The TRUST TRIANGLE in one call: composes reputation, standing (LAWBOR proven history, if BIII_LAWBOR_URL set) and settlement (on-chain, if amountMicro given) into ONE verdict (unsafe/unknown/trusted/settled). Reputation is shown as TWO LENSES kept SEPARATE, never merged: local = this node's known-bad screen against public lists (no network, decisive — a BLOCK overrides everything) and oracle = MainStreet's advisory read (ORACLE-REPORTED, can raise trust but never lower a local block). Fail-closed: absence is never trust; a local BLOCK holds even if the oracle is down. Every verdict carries the list's freshness (asOf/ageDays/stale).

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
railNooptional — the rail this payment settles on. BIII can WITNESS only "base" / "usdc-base" (it reads Base directly). Name any other rail — "card", "mastercard-agent-pay", "sepa", "stripe" — and the settlement vertex returns not_observable instead of pretending: the payment may well have completed, BIII simply cannot see it. Fail-closed: a rail we do not recognise is treated as unwitnessable, never as on-chain. Reputation and standing still decide, so a vetted counterparty stays payable; `proven` stays false because nothing was seen.
agentIdNooptional — the counterparty's ERC-8004 agentId; surfaces a SEPARATE, advisory, re-verifiable ERC-8004 reputation lens (interop, never rival)
amountMicroNooptional — if given, also check on-chain settlement to this address
resourceUrlNooptional — the endpoint/resource URL you would pay; enables the local phishing/plain-http/admin-path URL lens in the local classifier
counterpartyYes0x address to assess (the merchant, or a payer)
erc8004SummaryNooptional — a ReputationRegistry.getSummary result {count, summaryValue, summaryValueDecimals} for the agentId; BIII applies the sybil-honest lens + a re-verify pointer (BIII does not read the registry live yet)
erc8004TrustedClientsNooptional — attest the getSummary was filtered to clients YOU trust (drops the sybil caveat; still advisory)

Schema Changelog

Changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. Changed1 schema field changed
    • addedInput schema / properties / rail
      Added value: +{
+  "description": "optional — the rail this payment settles on. BIII can WITNESS only \"base\" / \"usdc-base\" (it reads Base directly). Name any other rail — \"card\", \"mastercard-agent-pay\", \"sepa\", \"stripe\" — and the settlement vertex returns not_observable instead of pretending: the payment may well have completed, BIII simply cannot see it. Fail-closed: a rail we do not recognise is treated as unwitnessable, never as on-chain. Reputation and standing still decide, so a vetted counterparty stays payable; `proven` stays false because nothing was seen.",
+  "type": "string"
+}
  2. First observed

TDQS

A4.1/5.0
Behavior4/5

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

With no annotations, the description carries full burden. It discloses key behaviors: fail-closed ('absence is never trust'), local block override ('BLOCK overrides everything'), and lens separation ('two lenses kept separate'). Missing explicit mention of idempotency or permissions.

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

Conciseness5/5

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

Description is approximately 100 words, front-loaded with the key concept, and every sentence adds value. No redundancy or unnecessary detail.

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

Completeness3/5

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

The description covers the tool's concept and behavioral traits well but lacks any specification of the return value structure (e.g., verdict object shape). Without an output schema, this omission is notable given the tool's complexity.

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?

Schema descriptions cover 100% of parameters. The tool description adds high-level context ('TRUST TRIANGLE', lens logic) that enriches understanding beyond individual parameter descriptions. A small deduction for not detailing the 'rail' parameter's behavior more explicitly in the main text.

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?

Description clearly states the tool 'composes reputation, standing, and settlement into ONE verdict' with explicit output values (unsafe/unknown/trusted/settled). The concept is unique among siblings (e.g., till_vet_agent focuses on agent-specific vetting), making its purpose distinct.

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

Usage Guidelines3/5

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

The description outlines the tool's inputs and logic (e.g., 'if amountMicro given' for settlement) but does not explicitly state when to prefer this tool over alternatives like till_vet_merchant. Usage context is implied but not formally defined.

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

All biii toolsInspect
Try in Browser
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

B3.1/5.0
Disambiguation3/5

Many tools have distinct, well-named purposes (vet_agent vs vet_merchant), but there is notable overlap between till_trust and till_vet_merchant (both provide trust verdicts), and till_launch_funder vs till_funder_history are closely related. The descriptions are detailed enough to differentiate, but an agent could still misselect between a few pairs.

Naming Consistency3/5

All tools share the till_ prefix, but the pattern is mixed: some use verb_noun (check_invoice, create_charge, watch_wallet) while others are noun phrases (key_exposure, open_approvals, rug_powers) or bare nouns (floor, trust, roll). This is readable but not predictable, so an agent cannot reliably guess a tool name from a verb.

Tool Count2/5

At 29 tools, this server exceeds the 25+ threshold that signals an overgrown toolkit. Even with a broad domain, many tools are one-off niche scanners (till_b20_authentic, till_floor, till_meter) that inflate the surface and could be consolidated or externalized.

Completeness4/5

The toolkit covers the payment lifecycle comprehensively: create charges/invoices, check payments, verify delivery, generate receipts, rolls, and accounting exports. It also spans identity, trust, security scanning, and theft tracing. Minor gaps exist (no update/cancel for charges, no token-general vetting), but the non-custodial, read-only design makes these acceptable.

Resources