Skip to main content
Glama
Sign In
physverify

physverify

by rcgiri-physics
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

confirm_claim

Runs a physics claim file in a fresh process to judge a suspected defect, returning confirmed, rejected, or not decidable with reasons and each check's status, then appends the result to a ledger.

Instructions

Judge one suspected defect with physverify confirm. Takes a claim file that defines CLAIM, control() and checks() with a reference written from the physics (docs/confirm.md), runs it in a fresh process and returns the verdict: confirmed, rejected or not decidable, with the reason and each check's status and message. Appends the result to a ledger file. Runs the code, so it needs trusted=true or the sandbox (PHYSVERIFY_REPO set).

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "claim": "<claim>"
    },
    "name": "confirm_claim"
  }
}

Input Schema

NameRequiredDescriptionDefault
claimYespath of the claim file
ledgerNoledger file (default confirm_ledger.jsonl)
timeoutNoseconds for the whole run (default 300)
trustedNothe code under test is trusted (required outside the sandbox)
allow_dependent_referenceNoreason, recorded, to run although the reference seems to use the code under test

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A4/5.0
Behavior5/5

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

With no annotations, the description carries the full burden and does it well: it says the code is run in a fresh process, that the result is appended to a ledger file (a real side effect), that execution requires trusted=true or a sandbox indicated by PHYSVERIFY_REPO, and it enumerates the verdict values and per-check output. The allow_dependent_reference escape hatch is also flagged as a recorded exception rather than a silent bypass.

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?

Four dense sentences, front-loaded with what the tool does, then mechanics, then return shape, then the safety gate. Every sentence carries information, though the parenthetical file references make it slightly cluttered to scan.

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?

For a five-parameter execution tool with no annotations and no output schema, the definition covers the return contents (verdict, reason, per-check status and message) and the side effect on the ledger. Minor gaps remain around how the ledger path is resolved and what distinguishes a 'not decidable' outcome, but nothing an agent needs in order to invoke it correctly is missing.

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 coverage is 100%, so the baseline is 3; the description still adds meaning by explaining what the claim file must contain (CLAIM, control(), checks(), a physics-derived reference, pointing to docs/confirm.md) and by tying trusted to the sandbox requirement. This goes beyond the terse schema strings, though ledger placement and timeout semantics are left to the schema.

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

Purpose4/5

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

States a specific verb and resource: it judges one suspected defect by running a claim file through physverify confirm and returning a verdict. The scope ('one suspected defect', 'fresh process') is concrete enough that an agent knows exactly what the tool does. It does not contrast itself with the sibling run_checks, which is the only missing element.

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?

Usage is implied by 'Judge one suspected defect' plus the prerequisite that the claim file must define CLAIM, control() and checks(). The trust/sandbox requirement is framed as a condition of execution, which helps. However, there is no explicit statement of when to prefer this over run_checks or when a claim is not worth confirming.

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