What it does
Catches your AI coding agent when it games the tests - locked-test hashes, tamper detection.
- Connects –
- Account or key No key needed
- Category 开发者工具
可搭配使用
教 AI 做这类工作的 Skill,可与本项一起使用
andrej-karpathy-skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
graphify
Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, every edge explained, no vector store.
为什么排在这里
列表按使用分(满分 100)排序,由以下部分组成
| 热度 | 0/80 | GitHub 星标或每周下载量:较大的计 60,较小的计 20 |
|---|---|---|
| 新鲜度 | 3/12 | 30 天内有发布或代码更新得满分 |
| 仍在响应 | 0/8 | 上次检查时云端端点有响应 |
| 使用分 | 3/100 |
质量: 基础检查通过 3/3 项(仅展示,不用于排序) · 上升势头: 暂无数据 (只用于“上升中”标签)
安装
选择你使用的 AI 应用,然后复制设置。
通过 3/3 项检查 · 在你的电脑上运行
第一步:你用哪个 AI?
选择你用的应用,步骤会显示在下方。
ChatGPT 暂时无法连接 Keel
开发者尚未在 MCP 注册表中填写安装方式
你可以改用这些方式
B · 开发者工具 类中能连接 ChatGPT 的其他服务器
从同类中使用较多的里挑选,仅限用地址即可连接、无需密钥的
- Sentry✓ ChatGPT
- Compiler Explorer✓ ChatGPT
- GitDiagram✓ ChatGPT
Claude Code 暂时无法连接 Keel
开发者尚未在 MCP 注册表中填写安装方式
Cursor 暂时无法连接 Keel
开发者尚未在 MCP 注册表中填写安装方式
VS Code 暂时无法连接 Keel
开发者尚未在 MCP 注册表中填写安装方式
一览
它能做什么
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
基础检查i
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月19日
- ✓ 通过
作者与仓库一致
注册名称已通过 GitHub 账户 tdalbanmoreynas 验证,该账户拥有此仓库
- ✓ 通过
仍在维护
代码最后更新于4个月前
- ✓ 通过
有许可证
使用 MIT
- – 不适用
端点有响应
它在你的电脑上运行,没有互联网端点
- – 不适用
使用 OAuth 登录
在你的电脑上运行,改用本地设置
- … 尚未检查
工具描述中没有隐藏指令
它在你的电脑上运行;我们只能读取互联网端点的工具列表
作者:申请重新检查或报告信息错误
给作者:在 README 中加徽章
如果你是作者,可把这段代码贴到 README。徽章会链接回本页。
[](https://mcp.popnx.com/en/servers/tdalbanmoreynas-keel)讨论
提问、分享技巧或报告无法使用。按“有帮助”数量排序。
用户评价
还没有评价
与“开发者工具”中的其他服务器比较i
- 星标数多于此分类中其他 6116 个服务器的 0%
- 发布到注册中心的时间比此分类中其他 6116 个服务器的 18% 更新
- 基础检查结果等于或优于此分类中其他 6116 个服务器的 99%
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
常见问题
Keel 安全吗?
它在 3 项适用的基础检查中通过了 3 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
它需要哪些密钥或代码?
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在我的电脑上运行还是在云端?
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
它还在维护吗?
GitHub 仓库最后更新于4个月前,最新版本(1.0.0)于4个月前进入注册中心。
哪些应用可以使用它?
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。
有什么可以替代它?
同一分类中有:GitDiagram, Appllama, Cloudflare MCP, Omnara, Repowise
功能相近的其他选择
同一分类、其他作者、命令已通过检查。官方的排在前面,再按星标排序。
如何看卡片:↓ 每周下载量i★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
GitDiagram
Architecture diagrams, explanations and Mermaid source for public GitHub repositories.
Appllama
Study screens, flows and paywalls from top-earning iOS apps, then build from what wins.
Omnara
Deploy durable AI agents with Omnara, an open-source managed-agent platform.
Repowise
Codebase intelligence for AI coding agents: graph, git history, docs, decisions, code health.
| 本项 Keel | GitDiagram | Appllama | Cloudflare MCP | Omnara | Repowise | |
|---|---|---|---|---|---|---|
| 运行位置 | 需要安装在电脑上 | 在云端运行,无需安装 | 在云端运行,无需安装 | 在云端运行,无需安装 | 在云端运行,无需安装 | 云端或安装在电脑上都可以 |
| 需要密钥 | 否 | 否 | 否 | 否 | 否 | 否 |
| 可在 Claude 中按 + 添加 | 否 | 否 | 否 | 是 | 否 | 否 |
| 基础检查 | 3/3 | 4/4 | 4/4 | 4/4 | 4/4 | 5/5 |
| 许可证 | MIT | MIT | MIT | Apache-2.0 | Apache-2.0 | AGPL-3.0 |
| 价格 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 |
作者的 README
打开 README
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
keel
An observable autonomy harness for AI coding agents. — *by TokenDrift*
keel is an MCP server that gives your coding agent (Claude Code, Cline, …) the tools to run a long task end to end — without drifting, cheating, or losing its work — while recording everything it does so you can steer on cost and quality.
You start the task. The agent disciplines itself. You come back to a milestone report.
The name: a keel is the part of a boat you never see — the spine under the waterline that stops it from being blown sideways by the wind. That's the job.
The problem
An agent left alone on a long task tends to do four things you can't see until it's too late:
- It loses the thread. Context gets compacted, earlier decisions are forgotten, work is redone or contradicted.
- It cheats. Faced with "make the tests pass," an agent will quietly edit the tests themselves — reward hacking. The suite goes green; nothing actually works.
- It loses files. An overwrite, a bad refactor, an hour of work gone with no checkpoint.
- It's a black box on cost. You have no idea what the run cost you, which model burned the budget, or whether the agent looped twenty times on one step.
keel puts a tool on each of these, and the agent calls them itself — because the rules are injected into its CLAUDE.md at init. Every call emits an event, so the whole run is legible afterward.
keel in action
A real run on a throwaway calc-kit project — the agent was told to implement add(a, b) so the locked tests pass, following the keel protocol. This is the actual sequence, unedited:
1. load_context → project loaded, tests intact
- record_decision → "add = a + b, minimal, no deps" logged
- (writes calc.py)
- save_progress → state persisted
- verify → green: 2 passed, tampered: false
- checkpoint → commit fac81a4 — milestone CERTIFIED
- check_drift → no drift
- log_usage → sonnet, 4000/1200 tok → $0.03
- route_task → (sensitive) recommends EU-hosted model; US + CN clouds excluded
- stats → 1 verify / 100% pass, 1 certified milestone, 0 cheat alerts
- report → milestone report written to .keel/reports/
Then, to show the anti-cheat is real — we modified a locked test file by hand (simulating an agent tampering to force a green) and ran verify:
{
"tampered": true, "tests_changed": ["~ tests/test_calc.py (modified)"], "passed": true }
tampered: true and passed: true at the same time. keel doesn't confuse "the tests pass" with "the tests are intact" — it reports both axes separately. An agent can make the suite green *by rigging it*, and keel still sees it. That distinction is the whole point.
The anti-cheat: a proof regime, not a prison
The hard problem with any optional-tool harness: an agent can ignore verify and just run pytest in the terminal, short-circuiting the one mechanism that catches cheating. A polite instruction in CLAUDE.md shapes what the agent *tries* to do — it doesn't *stop* anything.
keel closes this with three layers, on the principle that no single layer is perfect but together no path is simultaneously easy, silent, and untraced:
- Persuasion — the
CLAUDE.mdblock tells the agent, in no uncertain terms, thatverifyis the only judge and direct test runs are forbidden. Handles the dominant case: an agent that drifts out of negligence, not malice.
- A mechanical guard —
init_projectinstalls aPreToolUsehook (keel-guard) into the project's.claude/settings.json. It interceptspytest,python -m pytest,npm test,yarn test,go testbefore they execute and refuses them, redirecting the agent toverify. This runs in the Claude Code harness, outside the model's control. On our setup it holds even under `--dangerously-skip-permissions` (bypass mode skips interactive confirmations, not hooks — verify on your own machine, see *Honest limits*).
- Tamper-proof attestation — for anything that slips through,
verifyemits a hash of the exact working-tree state it judged, andcheckpointonly marks a milestonecertifiedif a passing, untamperedverifyran against *that same tree*. A bypass produces an uncertified milestone — visible instatsandreport— rather than a silent one.
Layer 2 makes the obvious bypass hard. Layer 3 makes *every* bypass visible. keel certifies what it witnessed; it doesn't claim to make evasion impossible — and that honesty is exactly what produces the "% certified milestones" signal worth tracking over time.
The golden rule that makes it work: you write the tests, never the agent. verify runs your locked suite and detects any change to it, file by file. If you change the tests yourself, re-run lock_tests (a human action).
Install
keel is plain Python — one dependency (the MCP SDK, which pulls in FastMCP and pydantic), plus git. It is developed and hardened on Windows, and runs on macOS/Linux.
git clone https://github.com/tdalbanmoreynas/keel.git
cd keel python -m venv .venv pip install -r requirements.txt python keel_server.py # must start without error (Ctrl-C to quit)
On Windows the venv activation is .venv\Scripts\activate; on macOS/Linux it's source .venv/bin/activate. Activation is optional if you point your client at the venv's Python directly (below).
Connect to Claude Code
# from the keel folder, pointing at the venv's interpreter (absolute paths)
macOS/Linux:
claude mcp add keel -- /absolute/path/to/keel/.venv/bin/python /absolute/path/to/keel/keel_server.py
Windows example:
claude mcp add keel -- "D:\path\to\keel\.venv\Scripts\python.exe" "D:\path\to\keel\keel_server.py"
Or paste the block from mcp_config.example.json into your client's config (Cline: cline_mcp_settings.json).
Two things that will save you a debugging session:
- Add an allow rule for keel's tools in your Claude Code settings (
mcp__keel__*), or the agent's first keel call hangs waiting on an authorization prompt that doesn't always surface. - Launch your client from the project root. The
keel-guardhook loads from the current directory's.claude/settings.jsonwith no parent fallback — it's active when Claude Code runs from where the project lives.
Compatibility
keel has two parts, and they have different reach.
The 12 MCP tools are standard MCP — they work with any MCP-capable coding agent: Claude Code, Codex, Cursor, Cline, Windsurf, Zed, VS Code + Copilot, Continue, Goose, and the Cline forks (Roo Code, Kilo Code). Memory, checkpoints, the anti-tamper verify, attestation, decision logging, cost tracking, and route_task all run anywhere MCP runs.
The `keel-guard` hook (layer 2) is Claude Code–specific. It's a PreToolUse hook in .claude/settings.json — a Claude Code mechanism that no other client implements today. So:
| Client | MCP tools (layers 1 & 3) | Mechanical guard (layer 2) |
|---|---|---|
| Claude Code | ✅ Full | ✅ Yes — the keel-guard hook installs and fires |
| Codex, Cursor, Cline, Windsurf, Zed, Copilot, Continue, Goose | ✅ Full | ⚠️ Not installed — no hook system |
On a non–Claude Code client you still get the persuasion layer (the injected rules) and, crucially, the attestation layer: a checkpoint that skipped verify still comes out uncertified in stats and report. You lose the mechanical block on direct test runs, but you don't lose visibility — the bypass is still recorded. That's the proof-regime design paying off: the guarantee that *matters* (you can always tell whether a milestone was verified) doesn't depend on any one client's hook system.
Claude Code is where keel is fullest and where it's developed and hardened. Other clients are first-class for everything except layer 2.
Workflow
- Once per project —
init_project: creates.keel/, writes the architecture and immutable constraints, locks your test files, initializes git, installs thekeel-guardhook, and injects the autonomy rules intoCLAUDE.md. - Then the agent self-manages by following those rules:
load_contextat the start of every session,save_progress+checkpointafter each step,verifyto validate,log_usageafter each model call,record_decisionat each architectural choice,reportat the end.
You come back at milestones, not at every line.
The 12 tools
| Tool | Role | Guards against |
|---|---|---|
init_project | Lays the harness (architecture, constraints, test lock, git, hook, CLAUDE.md) | — |
load_context | Resumption briefing at session start | Context loss |
save_progress | Writes done / doing / todo to disk | Context loss |
record_decision | Append-only log of architectural choices | Architectural drift |
checkpoint | Git commit of the step, with certification | File loss |
verify | Runs the locked tests, detects tampering | Cheating (reward hacking) |
lock_tests | Re-locks the tests (human action) | — |
check_drift | Recalls constraints, flags unsaved work | Drift |
route_task | Most cost-effective model for a subtask | Cost / data sovereignty |
log_usage | Records real spend of a model call | Cost observability |
stats | Aggregated metrics (cost, cheating, certification, velocity) | Steering |
report | Markdown milestone report for the human | Steering |
What keel records
Every tool emits an event to .keel/events.jsonl (local, private to your repo). From that trace, keel continuously derives: sessions, checkpoints, certified vs. uncertified milestones, verify pass rate, cheating alerts, real cost per model, and velocity. That trace is what makes an agentic run legible — and it's the raw material for the funded layer below.
Open-core — keel is the free sensor
keel is one product in two pieces. This repo — the local sensor — is free and MIT, forever. It runs entirely on your machine: it watches your agent, catches tampering, tracks waste and cost, and writes everything to a local events.jsonl that never leaves your repo. A solo developer needs nothing else.
The funded piece is a cloud dashboard for engineering teams (*by TokenDrift*), planned, on top of the same sensor. Where the free tier answers *"how is my agent doing?"*, the team dashboard answers *"how is my whole team doing, over time?"* — by aggregating the sensors of every developer on the team: total agent spend and recoverable waste, the share of milestones that were honestly certified (not just green), real-time alerts when a budget or a cheat-rate spikes, and anonymized benchmarking against other teams (*"your agents tamper 3× more than the median"*).
The split is deliberate: the sensor is free because a local binary is cloneable and can't be sold — its job is to be useful and to spread. The paid value lives in the team-level network and aggregated data, which a single local install structurally can't produce. Same model as Langfuse, PostHog, and Sentry: free and complete for the individual; paid when you become a team that needs the collective view.
Privacy is built into the split: the sensor sends metadata only — never your code or prompts. Cheat detection runs locally; only the verdict (certified / suspect) would ever leave the machine.
A note on route_task and prices
The model prices (keel_core.py, MODELS constant) are June 2026 estimates — replace them with your real rates. route_task accounts for agentic burn (cost = tokens × iterations × inefficiency) and excludes non-EU clouds when sensitive=True, which is why a sensitive subtask in the run above was routed to an EU-hosted model.