Skip to main content
Glama

plan_fixes

Read-onlyIdempotent

Plans verified dependency fixes that close a uv or Poetry project's action alerts on a temporary copy, returning a diff, lock command, version moves, and remaining alerts.

Instructions

Work out the change that closes a project's action alerts and check it with the project's own resolver (uv lock --dry-run, or Poetry) on a temporary copy. Returns the requirement edits as a unified diff, the lock command to run, every version the lock moves (downgrades included), the alerts left out and why, and whether raising requires-python would unblock more fixes. Use it after triage_alerts reports action alerts; only uv and Poetry projects get a plan (for others it explains why there is none). Read-only: never edits or relocks the project; show the user the plan and apply the diff only when they agree. Needs network access, and resolving can build source distributions and take a few minutes on large projects.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "path": "<path>"
    },
    "name": "plan_fixes"
  }
}

Input Schema

NameRequiredDescriptionDefault
pathYesAbsolute path of the project folder: the one holding pyproject.toml, uv.lock, poetry.lock, requirements*.txt or package.json.
include_filteredNoAlso plan fixes for filtered alerts (dev-only, test-only...). They need no action, but Dependabot keeps them open until the package moves.
raise_python_floorNouv projects: when a fix needs a newer Python than requires-python allows, raise requires-python in the plan (checked with the resolver). This drops support for older Pythons: ask the user first.

Output Schema

NameRequiredDescriptionDefault
okYesfalse when no plan could be made; see error.
diffNoUnified diff of the requirement edits; null when none are needed.
hintNoSet when raise_python_floor=true would let more fixes in.
planNoThe plan: status, tool (uv or poetry), command, note, included (packages the plan fixes, with target versions and edits), left_out (with why), lock_changes (from/to) and python_floor.
errorNo
notesNo
appliedNoAlways false: the project is never changed.
detailsNoThe end of depquiet's stderr.
projectNo
pull_request_descriptionNoThe plan as Markdown, ready for a PR body.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed7 schema fields changedv0.1.1
    • addedInput schema / properties / include_filtered / description
      Added value: +"Also plan fixes for filtered alerts (dev-only, test-only...). They need no action, but Dependabot keeps them open until the package moves."
    • addedInput schema / properties / path / description
      Added value: +"Absolute path of the project folder: the one holding pyproject.toml, uv.lock, poetry.lock, requirements*.txt or package.json."
    • addedInput schema / properties / raise_python_floor / description
      Added value: +"uv projects: when a fix needs a newer Python than requires-python allows, raise requires-python in the plan (checked with the resolver). This drops support for older Pythons: ask the user first."
    • removedOutput schema / additionalProperties
      Removed value: -true
    • addedOutput schema / properties
      Added value: +{
      +  "applied": {
      +    "description": "Always false: the project is never changed.",
      +    "title": "Applied",
      +    "type": "boolean"
      +  },
      +  "details": {
      +    "description": "The end of depquiet's stderr.",
      +    "title": "Details",
      +    "type": "string"
      +  },
      +  "diff": {
      +    "anyOf": [
      +      {
      +        "type": "string"
      +      },
      +      {
      +        "type": "null"
      +      }
      +    ],
      +    "description": "Unified diff of the requirement edits; null when none are needed.",
      +    "title": "Diff"
      +  },
      +  "error": {
      +    "title": "Error",
      +    "type": "string"
      +  },
      +  "hint": {
      +    "anyOf": [
      +      {
      +        "type": "string"
      +      },
      +      {
      +        "type": "null"
      +      }
      +    ],
      +    "description": "Set when raise_python_floor=true would let more fixes in.",
      +    "title": "Hint"
      +  },
      +  "notes": {
      +    "anyOf": [
      +      {
      +        "type": "string"
      +      },
      +      {
      +        "type": "null"
      +      }
      +    ],
      +    "title": "Notes"
      +  },
      +  "ok": {
      +    "description": "false when no plan could be made; see error.",
      +    "title": "Ok",
      +    "type": "boolean"
      +  },
      +  "plan": {
      +    "additionalProperties": true,
      +    "description": "The plan: status, tool (uv or poetry), command, note, included (packages the plan fixes, with target versions and edits), left_out (with why), lock_changes (from/to) and python_floor.",
      +    "title": "Plan",
      +    "type": "object"
      +  },
      +  "project": {
      +    "title": "Project",
      +    "type": "string"
      +  },
      +  "pull_request_description": {
      +    "anyOf": [
      +      {
      +        "type": "string"
      +      },
      +      {
      +        "type": "null"
      +      }
      +    ],
      +    "description": "The plan as Markdown, ready for a PR body.",
      +    "title": "Pull Request Description"
      +  }
      +}
    • addedOutput schema / required
      Added value: +[
      +  "ok"
      +]
    • changedOutput schema / title
      Previous value: -"plan_fixesDictOutput"New value: +"PlanResult"
  2. First observedv0.1.0

TDQS

A4.8/5.0
Behavior5/5

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

Annotations already declare readOnlyHint and idempotentHint, and the description reinforces with concrete meaning ('never edits or relocks the project', work happens on a temporary copy). It adds operational context annotations cannot express: network access is required, resolution may build source distributions, and large projects can take minutes.

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-loaded with the core action, then constraints, then output contents — a sensible order with no filler sentences. It is dense and on the long side for a three-parameter tool, with minor overlap between the read-only clause and the annotations, but every sentence carries actionable information.

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?

An output schema exists, yet the description still orients the agent on what the plan contains (diff, lock command, moved versions including downgrades, excluded alerts and reasons). Combined with the network and runtime warnings and the resolver restriction, an agent has everything needed to call it and to set user expectations.

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, but the description ties parameters to outcomes: it frames raise_python_floor as 'whether raising requires-python would unblock more fixes' and echoes include_filtered through 'the alerts left out and why'. It does not add format or edge-case detail for path, so it stops short of a 5.

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 and resource ('work out the change that closes a project's action alerts') and pins down the mechanism ('check it with the project's own resolver ... on a temporary copy'). It is clearly distinguishable from triage_alerts, which it explicitly follows rather than replaces.

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?

Gives an explicit trigger ('Use it after triage_alerts reports action alerts'), an explicit exclusion ('only uv and Poetry projects get a plan'), and states what happens otherwise ('it explains why there is none'). It also gates the write step: show the plan and apply the diff only when the user agrees.

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