Coderbuds

Get guidance drift

get-guidance-drift-tool

Read-onlyIdempotent

Has this repo's agent guidance file drifted from the repo it describes? A CLAUDE.md or AGENTS.md that lists the directories under app/Domain (or any other path) is making a claim that decays silently — the list was right when written, the code kept moving, and nothing fails when the two diverge. Agents read the stale list as truth, so a rotted guidance file is worse than none: it teaches the wrong shape of the codebase with full confidence.

For each connected repository this returns the guidance file found, when it last changed, and every directory whose listing has gone stale — how many entries the file documents, how many are actually there now, and examples of what is missing. Call it before trusting a guidance file, when onboarding an agent onto an unfamiliar repo, or when deciding whether a hand-maintained inventory should be replaced by a get-team-context call.

Only ENUMERATIONS are flagged. A file that mentions a path in passing is referencing it, not claiming an inventory, and is never reported. A directory is flagged only when the file already names several of its children (so it is holding a list) and is missing several more. Repos with no guidance file, BitBucket repos, and unreadable repos report insufficient_data — never a false drift. Archived repositories are skipped. Use get-repo-readiness for whether a guidance file exists at all; this tool judges whether the one that exists is still true.

Input Schema

Name	Required	Description	Default
`repository`	No	Optional "owner/name" (or bare repo name) to assess a single repository instead of every repository on the team.

Output Schema

Name	Required	Description
`team`	Yes	Team name.
`rollup`	Yes	Team rollup: repositories_assessed, repositories_drifted, and undocumented_entries (the total number of directory entries no guidance file mentions).
`repositories`	Yes	Per-repository drift: repository slug, guidance_file (CLAUDE.md/AGENTS.md), last_updated date and days_since_update, status (drifted\|current\|insufficient_data), drifted_directories (path, documented, present, undocumented, examples), and a recommendation. Most-drifted repositories first.

Schema Changelog

Changes observed during successful MCP inspections.

First observedSep 11, 2026

TDQS

A4.6/5.0

Behavior4/5

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

Annotations already cover read-only and idempotent, yet the description adds substantial behavior beyond them: only enumerations are flagged, passing mentions are never reported, and no-file/BitBucket/unreadable repos return insufficient_data rather than false drift, with archived repos skipped. These false-positive-avoidance rules are exactly the kind of context annotations cannot express.

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 leading question frames the purpose, then rationale, then guidance, then edge-case rules – front-loaded and logically ordered. Slightly long, but nearly every sentence carries distinct information (the 'worse than none' rationale is the only expendable flourish).

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 so return detail isn't strictly needed, yet the description still summarizes what's returned (file, last change, stale directories with entry counts and examples). Combined with the edge-case rules, nothing an agent needs to decide whether and how to call it 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% and the single optional parameter is documented there, but the description reinforces the default behavior ('For each connected repository') versus the single-repo override, adding context the bare schema doesn't emphasize.

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 opens with a specific question naming the resource (agent guidance file) and the operation (detecting drift). It clearly distinguishes itself from get-repo-readiness and get-team-context by scope, so an agent can tell the tools apart without opening schemas.

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?

It gives explicit call conditions (before trusting a guidance file, when onboarding an agent, when deciding whether to replace a hand-maintained inventory) and names the alternative get-repo-readiness for the 'does a file exist' question. When-to-use and when-not are both present.

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

All Coderbuds tools Inspect

Try in Browser

Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?