linear-strict

Overview Schema Source Related Servers Discussions Admin

comment

Destructive

Post a typed entry on a Linear ticket: evidence, correction, ask, answer, or closed_by. Records findings, patches, questions, and links work under another ticket.

Instructions

Post a typed comment. evidence: a finding, optionally with an Observed patch. correction: must carry the description patch that makes the description say the corrected thing. ask: adds an OPEN row to Open questions. answer: must name the row it closes (answers: "Q3") and flips it to ANSWERED with a link. closed_by: this ticket's work landed under another ticket; names it (closed_by), sets the Linear relation (relation "duplicate" or "fixed_there") and records it under Fix. Move the state separately with set_status.

Input Schema

TableJSON Schema

Name	Required	Description
`base`	No	Required with patch. description_sha of the description this patch was written against: from get_issue, or from your last write to this ticket. The write is refused, with what changed, if the description has moved since.
`body`	Yes	Comment text, without a header: the server writes the header from kind and drops a copy typed here. For ask, the first line becomes the question row.
`kind`	Yes
`issue`	Yes	Issue identifier (e.g. ENG-123) or UUID
`patch`	No	Description sections to change. Impact is the one plain-language line at the top saying who notices this work (mode replace). Observed lines must be "YYYY-MM-DD · <source: command, SHA or query> · <what it showed>". Done when lines must be checklist items ("- [ ] <check that proves it>"). Open questions changes only through comment kinds ask and answer.
`ask_to`	No	For kind ask: who should answer
`answers`	No	For kind answer: the question id, e.g. Q3
`relation`	No	For kind closed_by: duplicate (same problem) or fixed_there (a different ticket whose change fixed this one)
`closed_by`	No	For kind closed_by: the ticket that carried the work
`author_label`	No	Agent name for the comment header, e.g. the model. With a personal API key the header reads "<author_label> via <you>"; with an agent (app) token it defaults to the agent identity.

Output Schema

TableJSON Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedv0.1.1Oct 1, 2026

changedInput schema / properties / body / description

Previous value: -"Comment text. For ask, the first line becomes the question row."New value: +"Comment text, without a header: the server writes the header from kind and drops a copy typed here. For ask, the first line becomes the question row."

First observedv0.1.0Sep 27, 2026

TDQS

A4.2/5.0

Behavior4/5

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

Annotations only declare the safety profile (destructive, non-idempotent, closed-world); the description goes further by disclosing the side effects of each kind — ask opens a row in Open questions, answer flips it to ANSWERED with a link, closed_by writes a Linear relation and records under Fix, patches mutate description sections. It does not mention the base/description_sha concurrency refusal (that lives only in the schema) or duplicate-post behavior, which keeps it from 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.

Conciseness4/5

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

Purpose is front-loaded in the first three words, and each following clause is a compact kind:behavior pair with no filler. It reads as a dense colon-separated block rather than flowing prose, but nothing in it is redundant against the schema or annotations.

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 10-parameter, five-mode write tool with annotations and an output schema, the description covers the mode semantics and the sibling boundary that an agent needs. It leaves the return payload to the output schema (correct) and only omits failure/duplicate behavior that the schema's base note partially covers.

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 90%, so the baseline is 3, but the description adds cross-parameter semantics the schema cannot express: which patch/kinds pair with which field (evidence → optional Observed patch, correction → mandatory description patch, answer → answers id, closed_by → closed_by + relation). That mapping is genuine added meaning over the enum listing.

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?

Opens with a concrete verb+resource ("Post a typed comment") and then enumerates the five kinds, so an agent knows this is a polymorphic write endpoint rather than a generic note-taker. It also carves out scope against set_status ("Move the state separately"). The tool name "comment" alone is vague, but the description disambiguates it well.

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

Usage Guidelines5/5

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

Each kind carries its own condition of use and its prerequisite: correction "must carry the description patch", answer "must name the row it closes (answers: \"Q3\")", closed_by names the carrying ticket and sets the relation. It also states explicitly that state changes belong to set_status, which is a real exclusion. This is about as close to per-kind when-to-use guidance as a tool gets.

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