Jean by Melad Labs

check_invariants

Read-only

Decide whether a sequence is scientifically admissible: protein unchanged, forbidden restriction sites absent, reading frame intact, length/GC/CAI inside stated bounds. Every verdict names what broke, where, and the remedy. Run this after any design or optimisation step, and stop the chain when admissible is false — never report a designed sequence without it.

Input Schema

Name	Required	Description
`table`	No	NCBI translation table. Advanced.
`_files`	No	The project: record references shaped like _subject. Their records become library documents a tool names in sequences/parts/document; a trace or alignment is found by file name.
`max_gc`	No	maximum GC fraction (0-1).
`min_gc`	No	minimum GC fraction (0-1).
`min_cai`	No	minimum CAI (needs cai_organism). Advanced.
`_subject`	No	The record to act on: exactly one of sequence, accession (+source), sample (an id from list_samples) or file ({name, text} or {name, base64}); record_index picks one record of several.
`sequence`	No	the sequence to judge — defaults to the open record.
`gc_window`	No	judge every window of this many bases instead of the whole sequence — a synthesis is rejected on a local GC peak, not on the mean. In bp. Advanced.
`max_length`	No	maximum length [bp]. In bp. Advanced.
`min_length`	No	minimum length [bp]. In bp. Advanced.
`cai_organism`	No	host for the codon-adaptation check, e.g. sce. Advanced.
`forbid_sites`	No	enzyme names that must not cut, e.g. BsaI, BsmBI.
`require_stop`	No	with require_frame: the last codon must be a stop. Only applies when require_frame is true. Advanced.
`require_frame`	No	length divisible by 3 and no premature stop.
`require_start`	No	with require_frame: the first codon must be a start codon. Only applies when require_frame is true. Advanced.
`preserve_protein_of`	No	the ORIGINAL coding sequence (or its protein): the result must still translate to it.

Output Schema

Name	Required	Description
`failed`	Yes	how many invariants did not hold
`checked`	Yes	how many invariants were asked
`summary`	Yes	one line per failure: value, requirement, remedy
`verdicts`	Yes	per-invariant verdicts, with breach coordinates
`admissible`	Yes	every invariant was decided and held

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedOct 21, 2026

changedInput schema / properties / require_start / description

Previous value: -"with require_frame: the first codon must be a start codon. Advanced."New value: +"with require_frame: the first codon must be a start codon. Only applies when require_frame is true. Advanced."

changedInput schema / properties / require_stop / description

Previous value: -"with require_frame: the last codon must be a stop. Advanced."New value: +"with require_frame: the last codon must be a stop. Only applies when require_frame is true. Advanced."

First observedOct 20, 2026

TDQS

A4.2/5.0

Behavior4/5

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

Annotations already declare readOnlyHint=true, destructiveHint=false and openWorldHint=false, so the safety profile is covered by structured data. The description adds genuinely new behavior beyond that: every verdict names what broke, where, and the remedy, and a false verdict is meant to halt the pipeline. It does not mention cost, speed, or size limits, keeping it short of a 5.

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?

Three sentences, front-loaded with the admissibility semantics, then verdict behavior, then the workflow obligation. Every sentence carries distinct information and none is padding.

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?

With an output schema present, the description is not obliged to explain return values, and for a 16-parameter nesting-heavy validator it covers purpose, verdict semantics, and workflow placement well. The one gap is that it never addresses how the many optional 'advanced' constraints default or interrelate, which an agent could want before wiring up parameters.

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?

Schema description coverage is 100%, so the schema already documents all 16 parameters (including the advanced GC-window, CAI-organism and frame flags). The description only restates the check categories at a conceptual level and adds no syntax, defaults, or format detail beyond the schema, so the 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 names a specific verb (decide/check) and resource (a sequence) and enumerates the exact admission criteria — protein preservation, forbidden restriction sites, frame, length/GC/CAI bounds — which maps tightly to this tool's parameter set and separates it from generic siblings like check_manufacturability or verify_construct. An agent knows precisely what class of question this tool answers.

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 gives an explicit trigger ('run this after any design or optimisation step') plus a workflow rule ('stop the chain when admissible is false') and a default use ('never report a designed sequence without it'). That is clear when-to-use guidance, though it names no alternative tool for the cases where a lighter-weight check would do.

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

All Jean by Melad Labs tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?