Skip to main content
Glama

Ask BRYDGE before acting

brydge_supervise
Idempotent

Request permission before executing an action (e.g., refund). Returns an authorization ID to record, or escalates to a person when rules need review.

Instructions

Ask BRYDGE whether you may take an action, before you take it. ALLOWED returns an authorization id: carry the action out and write that id into the record it creates, where the destination keeps it (for a Stripe refund: metadata.brydge_authorization). ESCALATED means a person decides: do not carry it out.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
factsNoWhat BRYDGE's rules judge the action by, such as {"amount": 4200}. BRYDGE compares a fact named amount with the amount in the destination's record, so give it in the same units.
actionYesThe kind of work, as it is set up in BRYDGE, such as refund.
targetYesWhat the work acts on: the charge, order or ticket id.
idempotency_keyYesA name for this one intended action, such as refund:ch_123. Send the same value if you retry the same action; never reuse it for a different one.

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
becauseYes
decisionYes
replayedYes
unobservedYes
authorizationYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4.2/5.0
Behavior4/5

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

Annotations already signal non-read-only, non-destructive, and idempotent behavior. The description adds valuable context: the ALLOWED response includes an authorization id to be written into the record, and ESCALATED means a human must decide. This goes beyond what annotations provide.

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?

Two tight sentences with no filler. The purpose is front-loaded, and the response-handling instructions are concise and actionable.

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

Completeness4/5

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

Given the output schema exists and idempotency is declared, the description covers the essential use case: when to call, what the response means, and how to proceed. It omits edge cases like denial handling beyond ESCALATED, but this is minor.

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

Parameters3/5

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

The schema provides 100% coverage with detailed descriptions for all four parameters, so the tool description doesn't need to add parameter info. It adds none, which is acceptable given the schema's richness; baseline 3 applies.

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 ('Ask'), a resource ('BRYDGE'), and the exact purpose ('whether you may take an action'). It clearly distinguishes this from sibling tools like brydge_verify or brydge_report_outcome by framing it as a pre-action permission gate.

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?

The description clearly says to use it 'before you take it' and explains how to handle ALLOWED vs ESCALATED outcomes. However, it does not explicitly name alternatives or state when NOT to use it, though the purpose is distinct enough that an agent can infer.

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