Skip to main content
Glama
Sign In
Mushi-Mushi

Mushi-Mushi

by kensaurus
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Remote

Preview CI cost fixes for every app

preview_ci_cost_fix_rollout
Read-onlyIdempotent

Preview grouped CI cost autofixes per app and repo, so each repository gets one draft pull request for all its fixes. Read-only; no changes are made.

Instructions

Preview 'Apply to every app' for the CI cost fixes (ADR 0047 amendment): every CI advice item that carries an autofix, across the apps this key may change (viewers' apps are left out), grouped per app and repo, so each repo can get ONE draft pull request (one CI run) for all its fixes. Per repo { projectId, projectName, repo, items [{ key, workflowPath, kind (add_concurrency | skip_drafts | weekly_cron | ignore_docs_paths), title, evidence, monthSavingUsd (estimated USD per 30 days, null when Mushi cannot price it), state (ready | pr_open | in_progress), prUrl }], readyItems, monthSavingUsd (the ready items' priced saving), unpricedItems, block (set when Mushi already knows the repo's GitHub access cannot edit workflow files: GITHUB_WORKFLOWS_WRITE_MISSING, from the setup check or the last attempt) }, largest saving first, plus totals, caps { reposPerRequest: 10, itemsPerRepo: 20 } and partial (set when part of the advice could not be read). Optional projectId limits it to one app; an account-level key with none reads every app. Read-only: it changes nothing. Open them with apply_ci_cost_fix_rollout.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "preview_ci_cost_fix_rollout"
  }
}

Input Schema

NameRequiredDescriptionDefault
projectIdNoLimit it to one app; omitted with an account-level key, every app you may change (`project_id` is accepted too.)

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv0.1.19

TDQS

A4.6/5.0
Behavior5/5

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

Beyond the readOnlyHint, the description discloses important behavior: viewers' apps are excluded, results are grouped per app and repo, ordering is largest saving first, caps are 10 repos and 20 items per repo, and a partial flag appears when advice cannot be read. It also names the block condition GITHUB_WORKFLOWS_WRITE_MISSING and explains that each repo maps to one draft PR and one CI run.

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?

The description is front-loaded with the purpose and the apply alternative, and the dense output-shape details are relevant because there is no output schema. It is long and the first sentence is heavily packed, but most content earns its place given the complexity.

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 no output schema, the description carries the full burden of explaining the return shape, and it does so thoroughly: per-repo fields, item fields, totals, caps, partial, and block reasons are all covered. Combined with its scoping and read-only behavior, an agent has enough to call it correctly.

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?

The schema already documents projectId at full coverage, but the description adds useful semantics: omitting it with an account-level key reads every app, while including it limits the preview to one app, and viewers' apps are still excluded. This goes slightly beyond the schema text.

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 states a specific verb and resource: previewing CI cost fix rollout across every app. It clearly distinguishes this read-only preview from the apply action by naming apply_ci_cost_fix_rollout and by explaining the grouping and per-repo draft PR purpose.

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 explicitly frames the tool as previewing 'Apply to every app' and tells the agent to open results with apply_ci_cost_fix_rollout, which is a clear next-step alternative. It also explains the projectId scoping choice, but it does not explicitly state when not to use this tool versus related CI cost tools such as get_ci_cost_fix_bundle or open_ci_cost_fix.

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