Jean — Molecular Biology & Genomics

design_sequencing_primers

Design a WALK of sequencing primers that reads a whole region, not one pair: tiled reads with confirmed overlap, on both strands by default, plus the coverage the walk actually achieves and any stretch it cannot reach.

Input Schema

Name	Required	Description
`_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.
`overlap`	No	confirmed overlap required between consecutive reads on one strand. Overlap is what turns abutting reads into a contig. Default 100 has NO published source and is a judgement, which is why it is stated rather than dressed as a figure. In bp.
`_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.
`dead_zone`	No	bases at the start of a read that are not trustworthy. Default 35 is the LONGEST the same source warns about (it says the first 25-35 bases are poor), so an untouched walk is the cautious reading in both directions. In bp. Advanced.
`target_tm`	No	target melting temperature for each primer in the walk. In °C. Advanced.
`min_length`	No	shortest acceptable primer. In bp. Advanced.
`region_end`	No	0-based inclusive end of the region to read. Unset = the whole template. Advanced.
`read_length`	No	usable bases per read, AFTER the dead zone. Default 800 is the shortest read MSU RTSF Genomics Core, Sanger Sequencing Best and Worst Practices (2024-04-25), ABI 3730XL promises (it quotes 800-900), which is the cautious end of a published range rather than a round number. In bp.
`both_strands`	No	read every base from both directions - what a sequencing facility means by double-stranded coverage, and the standard for anything deposited or published.
`region_start`	No	0-based inclusive start of the region to read. Unset = the whole template. Advanced.

Output Schema

Name	Required	Description
`gaps`	Yes	stretches no read reaches - empty means the region is fully read
`used`	Yes	the parameters the plan was actually built under, so a readout never restates them
`reads`	Yes	one entry per primer in the walk: name (F1, R1, ...), sequence, strand, and the stretch it is expected to read
`primers`	Yes	just the primer sequences, in walk order, for ordering
`skipped`	Yes	anchors where no primer could be designed, each with the reason - never a silently short walk
`complete`	Yes	true when every base of the region is read at least once
`coverage`	Yes	every stretch of the region, contiguous and in order, with its depth
`authority`	Yes	the published source the read-length and dead-zone defaults come from
`readCount`	Yes	how many primers the walk needs
`doubleStranded`	Yes	true when every base is read from both directions
`singleStranded`	Yes	stretches read from one strand only, when both were asked for

Schema Changelog

Changes observed during successful MCP inspections.

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 are minimal (readOnlyHint=false, destructiveHint=false, openWorldHint=false) and mostly cover safety. The description adds real behavioral context: reads are tiled with a confirmed overlap constraint, both strands is the default, and the result reports the coverage actually achieved plus unreachable stretches — telling the agent the output may be partial rather than a clean success.

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?

One dense sentence, front-loaded with the core action and scoped by the 'not one pair' contrast before the detail clauses. It is information-dense rather than padded; the trailing clause about coverage and unreachable stretches does earn its place by setting output expectations.

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?

With 10 parameters, a nested _subject object, and an output schema, the description covers the essential shape: it designs a walk, the walk is tiled and double-stranded by default, and the return can include gaps. Combined with the fully documented schema and existing output schema, an agent has what it needs to call this correctly.

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% and the parameter docs are unusually explicit (overlap, read_length, dead_zone all carry rationale and units), so the baseline is 3. The description only echoes the overlap and both-strands defaults rather than adding new meaning such as how region_start/region_end interact with the whole-template default.

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 gives a specific verb (design) and resource (sequencing primers) and immediately scopes it: a tiled walk over a whole region, explicitly 'not one pair'. That phrase distinguishes it from the many pair-design siblings such as design_primer_pairs and design_primers without the agent needing to open either schema.

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?

Usage is clear from the framing: pick this when the goal is to read an entire region via tiled reads rather than a single amplicon. It does not name a specific sibling alternative or state when *not* to use it, so the routing decision must still be inferred from the 'whole region vs one pair' contrast.

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

All Jean — Molecular Biology & Genomics 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?