Kosmodrom

task_create

Creates a task (not a bug) as a draft; the token owner reviews and publishes it. rulesRevision is the current revision of the workspace's task creation template for the type, and spec is checked against that template; placement is one of featureID / featureKey (product and version from the feature), versionID (no suitable feature) or productID (product backlog). Returns link, status and available transitions; without ord the task goes to the end of the feature plan.

Input Schema

TableJSON Schema

Name	Required	Description
`ord`	No	Position of the new bug; tasks are ordered by `ord` descending. Omitted - end of the feature's plan, so bugs created one by one keep their order.
`gist`	Yes	Gist: one or two sentences, "problem → solution", up to 300 characters of plain text (no Markdown, links or task keys), in the language of the workspace data.
`spec`	No	Spec (requirements) in the language of the workspace data, per the task creation template. Markdown (HTML is rejected); a write replaces the whole field, including `::: doc` sections and `attach:`, `design:`, `color:` tokens
`type`	Yes	Type by the nature of the work: planning, analysis, design, research, education, documentation, other - general tasks; frontend, backend, fullstack, devops, tests - development (client and server together - one fullstack task); freetest, smoke, regression, testcase - testing
`title`	Yes	Task title in the language of the workspace data, up to 200 characters, conveying the substance of the work rather than an area: "Authorization", "List", "Settings" are poor titles that don't distinguish the task
`featureID`	No	Feature ID - the preferred placement
`moduleIDs`	No	Product modules the task affects
`productID`	No	Product ID - the product backlog, last resort
`versionID`	No	Version ID - when no feature fits
`featureKey`	No	Feature key #F12 (or F12, 12); not with featureID
`componentIDs`	No	Product components the work actually touches; a module does not imply them. May stay empty if unsure
`testAssignee`	No	Tester: employee uuid, "any" (any employee takes the stage) or null (no stage); default - no testing
`workAssignee`	No	Assignee: employee UUID, "any" (any employee) or null (no work stage); default "any".
`workTimePlan`	No	Planned work time in minutes, per the assignee's role level
`rulesRevision`	Yes	Revision of the task creation template for this type; a stale one is rejected
`reviewAssignee`	No	Reviewer: employee UUID, "any" (any employee) or null (no review stage); default: no review.
`approvalAssignee`	No	Approver: employee UUID, "any" (any employee) or null (no approval stage); default: no approval.
`workIterationKey`	No	Iteration key for work outside the current iteration; default - current

Schema Changelog

Changes observed during successful MCP inspections.

First observedOct 6, 2026

TDQS

A4.3/5.0

Behavior4/5

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

Annotations only declare a non-destructive write; the description adds real behavioral context beyond that: the task lands as a draft that the token owner must review and publish, a stale rulesRevision is rejected, and it names the returned link/status/transitions. Auth/permission requirements are not covered, keeping this below 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?

Front-loads the purpose and packs placement, validation, and return info into one dense paragraph. At 18 parameters some density is justified, but the run-on clause chain costs readability versus short scannable sentences.

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 complex 18-parameter tool with no output schema, the description usefully covers the return surface, the template-revision requirement, ord semantics, and the draft/publish workflow. It stops short of failure modes and permission requirements, but nothing essential for a correct call 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 goes further by explaining cross-parameter placement logic (which key to choose and why) and the ord default behavior ('without ord the task goes to the end of the feature plan'), which the schema states only tersely.

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?

States a specific verb+resource ('Creates a task') and immediately scopes it against the sibling it is not ('not a bug'), which is exactly the distinction an agent needs versus bug_create. The draft/review lifecycle is also stated up front.

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?

Gives explicit routing among placement options: featureID/featureKey preferred, versionID when no feature fits, productID as last resort. It does not explicitly name bug_create as the alternative for bugs (only the parenthetical 'not a bug'), so exclusions are implied rather than spelled out.

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

All Kosmodrom 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?