whats-inherited-mcp
The whats-inherited-mcp server enumerates everything in a code checkout that is addressed to an AI agent — instruction files, hooks, declared MCP servers, and shipped agent extensions — so you can review the agent-facing surface of code you clone or PRs that touch agent config.
inherited_summary— Get a headline count of all agent-targeted items in a directory: instruction files, hook commands, declared MCP servers, and skills/commands/subagents. The recommended starting point before working in an unfamiliar repo or reviewing a PR that touches agent config.instruction_files— List every instruction file (e.g.,CLAUDE.md,AGENTS.md,.cursorrules) including nested ones deep in subdirectories, with size, estimated token cost, and what each@importline pulls in (including imports that resolve outside the repo). File contents are never returned — you open the paths yourself.auto_run_commands— Show hook commands declared in.claude/settings.json/settings.local.json: the command, its trigger (tool use, session start, prompt submit), and whether the referenced script lives inside the repo, outside it, or is missing.declared_mcp_servers— List MCP servers the repo declares (.mcp.jsonand equivalents): launch command, whether it fetches remote code from a package registry at start time, filesystem paths granted outside the checkout, and the names of environment variables it expects (values are never read).agent_extensions— Identify skills, slash commands, subagent definitions, and plugin markers shipped in the checkout's.claude/or.agents/directories (root, or 1–2 levels deep for monorepos), with counts, names, and sizes — never file contents.
All tools take an optional dir argument (absolute path); when omitted, they fall back to WI_DEFAULT_ROOT if set, otherwise the server's working directory. Conservative by design: no shell, no child processes, no network, no writes; file contents are never emitted, env var values are never read, and output is fenced to prevent injection. Reports what files declare, not what your client will actually do, and handles edge cases like symlinks, deduplication by real path, and truncation on large monorepos.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@whats-inherited-mcp What does this repo tell my agent to do?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
whats-inherited-mcp
You review the code you clone. Almost nobody reviews the part of it that talks to your agent. An MCP server that enumerates everything in a checkout addressed to an AI agent rather than to you: instruction files, hook commands wired to agent events, MCP servers the repo declares, and the skills and subagents it ships.
Why
git diff shows you code, and you read code. It also shows you three added lines in a CLAUDE.md, and you skim those, because they look like documentation. They are not documentation — they are instructions your model will follow.
The surface is bigger than most people picture. A directory you cloned can carry:
CLAUDE.md/AGENTS.md/.cursorrules— loaded into context and treated as instructions, including nested copies deep in the tree that only apply when the agent works in that subdirectoryhook commands in
.claude/settings.json— shell wired to fire on tool use, session start, or prompt submit.mcp.json— MCP servers the repo asks to add, often launched withnpx -y <package>, which means the code that runs is downloaded at start time and is not the code you reviewed.claude/skills,.claude/commands,.claude/agents— capabilities the repo hands the agent
Nothing collects that in one place. This does.
Run against a checkout of langfuse/langfuse at 7d2afa4 — an ordinary, reputable open-source repo, picked precisely because there is nothing wrong with it:
# Inherited agent surface
**12 item(s) in this checkout are addressed to an agent, not to you.**
| Surface | Count | Detail |
|-------------------------------|------:|--------------------------------------------------------------|
| Instruction files | 12 | ~41,848 est. tokens, 5,593 lines your agent is told to follow |
| Hook commands | 0 | configured to run on agent events |
| MCP servers declared | 0 | 0 fetch code from a registry at launch |
| Skills / commands / subagents | 33 extensions (196 files) | shipped under `.agents/`, available to the agent |
## Worth a look
- 11 instruction file(s) are **not at the repo root** — they apply when the agent
works in those subdirectories and are easy to miss in review.and instruction_files adds:
> Counted once, reachable under more than one name (symlinks):
> - `AGENTS.md` ← also `.agents/AGENTS.md`, `CLAUDE.md`Five and a half thousand lines of standing instruction, most of it in files you would never open, in a repo nobody has any reason to distrust. That is the point: the number is large even in the benign case, which is exactly why an unusual entry in it goes unnoticed.
Related MCP server: repo-ready
Tools
Tool | What it answers |
| The headline: everything in this checkout addressed to an agent. Start here |
| Every |
| Hook commands the checkout wires to agent events, and whether the script each references is inside the repo, outside it, or missing |
| MCP servers the repo declares, which of them fetch code at launch, and filesystem paths they are granted outside the checkout |
| Skills, slash commands and subagents the repo ships |
Every tool takes an optional dir. When it is omitted the server falls back to
WI_DEFAULT_ROOT if that is set, and otherwise to its working directory.
Install
Claude Desktop (one-click, no terminal): download the latest whats-inherited-mcp-<version>.dxt from Releases and open it with Claude Desktop (double-click, or Settings → Extensions → Install Extension…). The server and its dependencies ship inside the bundle — no npm, no Node install.
Register with Claude Code (available in every session):
claude mcp add --scope user whats-inherited -- npx -y whats-inherited-mcpOr in any MCP client config:
{
"mcpServers": {
"whats-inherited": {
"command": "npx",
"args": ["-y", "whats-inherited-mcp"]
}
}
}git clone https://github.com/stcmain/whats-inherited-mcp.git
cd whats-inherited-mcp
npm install && npm run build
# then point your client at node /path/to/whats-inherited-mcp/dist/index.jsPublished as whats-inherited-mcp on npm and as
io.github.stcmain/whats-inherited-mcp in the MCP Registry.
Configuration
One optional setting, and it takes no credentials.
Variable | Default | Meaning |
| the server's working directory | Directory to inspect when a tool is called without a |
Every tool accepts an explicit dir, which always wins. WI_DEFAULT_ROOT only
changes the fallback, and it is worth setting when a desktop client launches the
server: the process then inherits that client's working directory, which is
rarely the checkout you meant to inspect.
{
"mcpServers": {
"whats-inherited": {
"command": "npx",
"args": ["-y", "whats-inherited-mcp"],
"env": { "WI_DEFAULT_ROOT": "/path/to/the/checkout" }
}
}
}What it counts, and what it refuses to guess
Inflating this in the alarming direction would be easy and would make the tool useless, so the accounting is deliberately conservative:
It does not detect malicious content. There is no heuristic scanner, no "suspicious phrase" regex, no risk score. Those produce confident false positives on ordinary repos and miss anything written with care. This server tells you where to look; you do the reading.
Files are counted once. A repo can expose one file under several names —
CLAUDE.md → AGENTS.md → .agents/AGENTS.mdis a real pattern in the wild. Entries are deduplicated by resolved real path and the aliases are listed, rather than counting the same content three times..claude/is not double-counted. A skill's ownCLAUDE.mdis reported as a skill, not also as a project instruction file."No path token identified" is not a safety claim. When a hook command has no filesystem path this server can confidently extract, it says so and stops. That is a stated gap in the analysis, not a verdict.
Import detection is conservative. Fenced code blocks are stripped first, and an unrooted
@tokenonly counts when it names a document — so@scope/pkgand@mentionsstay out of the number.
Honest limitations
It reports; it does not judge, and it does not fix. Nothing is edited, quarantined or scored. Every item it lists is normal in a legitimate repo.
Whether your client actually runs project hooks is your client's business. Clients differ, and they prompt differently and change between versions. This server reports what the files declare, not what your client will do with them.
Token counts are estimates (~4 chars/token). Treat them as a ranking and a rough scale, not as billing. Anthropic's tokenizer is not public, so nothing local can do better.
Claude Code layout is the model. Cursor, Windsurf, Cline and Copilot instruction files are recognised, but hook and MCP parsing follows the Claude Code schema.
Very large monorepos are truncated. The walk is depth- and entry-capped; when the cap is hit the output says so and marks the results partial rather than quietly under-reporting.
It never reads git history. It describes the working tree as it is on disk right now, not what a diff changed.
Symlinked directories are not followed (loop risk). Symlinked files are.
Design notes / threat model
This server's whole job is to look at content that may be hostile, so the design assumes it is.
It must not become the injection vector it reports on. The body of an instruction file is never returned — only metadata, paths and structured fields parsed out of known JSON config keys. Pasting a repo's
CLAUDE.mdinto your context to tell you the repo might contain something bad would be self-defeating.Repo-authored strings are fenced and labelled. Hook commands and MCP launch lines have to be shown to be useful. They are emitted inside inline code spans with backticks neutralised, pipes escaped and newlines flattened so a crafted string cannot break out of the span or out of a markdown table, and every block carries a standing note that the quoted text is data from the checkout, not instructions.
No child processes. No shell. No network. No writes. The only Node APIs used are
node:fsreads,node:pathandnode:os. There is nochild_processimport anywhere in the source, so nothing in a scanned repo can be executed by scanning it.diris the one model-controlled path, and it is bounded by construction: it is resolved, real-pathed and required to be an existing directory. Because file bodies are never emitted, pointing it somewhere sensitive discloses filenames and sizes, never contents — and it cannot write, execute or transmit anything.Environment variable values are never read — only names.
.mcp.jsonis a place people leave API keys in plaintext.Bounded work: depth cap, entry cap, file-size ceiling, and no symlinked-directory traversal.
Who makes this
Built by Shift The Culture — we run a one-person company on AI agents and ship the tooling we needed ourselves. This server is free and MIT-licensed, no strings.
It has three siblings, all also free and MIT:
whats-running-mcp — what is actually running on the box right now, instead of what an old transcript claims.
whats-loaded-mcp — what is eating your context window before you type: skill descriptions, memory files and their imports.
whats-allowed-mcp — what your agent can do without asking you: merged permission rules, which settings file wins, and the rules your client accepts and then ignores.
The rest of that tooling is paid:
Agent Fleet Ops Kit ($29) — the other failure modes of running three or four agents on one box: two sessions editing the same checkout, a dev server nobody owns (so the agent tests a different app than it edits), and MCP servers leaked from crashed sessions that hold ports and RAM for weeks. Prefer PayPal? Same kit on Payhip.
Agent Reliability Kit ($29) — a Stop hook and two CLIs that block a turn when an agent claims "done" against a repo, URL, or build that was never actually checked. Prefer PayPal? Same kit on Payhip.
The server above stays free and MIT either way — it has no upsell in it, no telemetry, and no dependency on the paid kits.
Sponsors
This server is MIT and stays MIT. There is no pro edition, no telemetry, and nothing held back from the free build. Sponsorship is how the maintenance gets paid for without any of that changing.
No sponsors yet — the first slot is open. Company sponsors get their name or logo in this section, in the two sibling servers, and on the sponsor page. Tiers, exactly what the placement is, and what it explicitly does not buy: https://shifttheculture.media/sponsor
Individuals: https://paypal.me/ShiftTheCultureLLC — any amount, no perks, no tier.
License
MIT © Zachary Pampu
Available Tools
5 toolsagent_extensionsSkills, commands and subagents the repo shipsA
Skills, slash commands, subagent definitions and plugin markers shipped inside the checkout's .claude/ or .agents/ directories (root, or one to two levels down — monorepos put them per package). These become available to an agent working in this directory. Counts distinct extensions and the markdown files they carry. Names and sizes only — contents are never returned.
Inputs: dir.
auto_run_commandsCommands this checkout wires to agent eventsA
Hook commands declared in the checkout's .claude/settings.json and settings.local.json — shell commands wired to fire on agent events such as tool use, session start or prompt submit. Shows the command, its trigger, and whether the script it references lives inside the repo, outside it, or is missing.
Inputs: dir.
declared_mcp_serversMCP servers this checkout declaresA
MCP servers declared by the checkout (.mcp.json and equivalents) — the launch command, whether it fetches code from a package registry at start time, filesystem paths it is granted outside the repo, and the NAMES of environment variables it expects. Values are never read.
Inputs: dir.
inherited_summaryWhat this checkout hands your agentA
Start here. Headline count of everything in a directory that is addressed to an AI agent rather than to you: instruction files, hook commands wired to run automatically, MCP servers the repo declares, and skills/commands/subagents it ships. Use before working in a repo you did not write, or when reviewing a PR that touches agent config.
Inputs: dir.
instruction_filesFiles that instruct the agentA
Every CLAUDE.md, AGENTS.md, .cursorrules and equivalent in the checkout, with size, estimated token cost, and what each one's @import lines pull in — including imports that resolve outside the repo. File contents are never returned; open the paths yourself.
Inputs: dir.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 5 tools
Each tool targets a distinct category of AI-facing repo content: MCP servers, instruction files, hook commands, agent extensions, and a summary count. There is no overlap in what they return.
All tool names follow a consistent lower_snake_case noun-phrase pattern describing the resource they expose. No verb confusion or mixed conventions.
Five tools is well-scoped for inspecting AI-related repository configuration. Each tool earns its place without redundancy or excessive granularity.
The set covers the major categories of AI-agent-related configuration: declared MCP servers, instruction files, auto-run hooks, agent extensions, and a summary. No obvious missing operation for the stated purpose.
Maintenance
Related MCP Connectors
Free, source-checked directory of AI agent tools, MCP servers and skills, with read-only search.
Scan AI agent skills, MCP server packages and MCP configs before you install them. Static analysis.
Find llms.txt, docs, MCP servers, skills and CLIs for a tech stack, for coding agents.
Scan AI agent skills and configs for hidden Unicode, prompt injection and exfiltration.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceScans projects to discover all AI configuration files, MCP servers, agents, and memories across 19 tools, enabling export and import of the entire AI ecosystem.93 npm1MIT
- AlicenseNot gradedqualityCmaintenanceExposes tools that scan a repository's manifests, CI workflows and layout to generate truthful AGENTS.md/CLAUDE.md agent instructions, then audit those docs for drift such as dead commands, undocumented scripts and CI mismatches. Enables agents to keep their own onboarding documentation accurate and CI-verifiable instead of hand-written and rotting.1MIT
- AlicenseNot gradedqualityBmaintenanceA read-only MCP server that lets agents scan a project's AI agent setup — Claude Code, Codex, Cursor, MCP servers and instruction files — and surface security findings such as the lethal trifecta, broad permissions, exposed secrets and unpinned packages.57 npmMIT
- AlicenseNot gradedqualityBmaintenanceEnables agents to check their own instruction files for drift before a run, returning findings with file and line for stale counts, missing files, broken links, and missing commands. Also lets agents list which agent docs get inspected.14 npmMIT