ingero 是 AI Skill Hub 本期精选MCP工具之一。综合评分 7.1 分,整体质量较高。我们推荐使用将其纳入你的 AI 工具库,帮助提升工作效率。
ingero 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。
ingero 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。
# 方式一:通过 Claude Code CLI 一键安装
claude skill install https://github.com/ingero-io/ingero
# 方式二:手动配置 claude_desktop_config.json
{
"mcpServers": {
"ingero": {
"command": "npx",
"args": ["-y", "ingero"]
}
}
}
# 配置文件位置
# macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
# Windows: %APPDATA%/Claude/claude_desktop_config.json
# 安装后在 Claude 对话中直接使用 # 示例: 用户: 请帮我用 ingero 执行以下任务... Claude: [自动调用 ingero MCP 工具处理请求] # 查看可用工具列表 # 在 Claude 中输入:"列出所有可用的 MCP 工具"
// claude_desktop_config.json 配置示例
{
"mcpServers": {
"ingero": {
"command": "npx",
"args": ["-y", "ingero"],
"env": {
// "API_KEY": "your-api-key-here"
}
}
}
}
// 保存后重启 Claude Desktop 生效
Featured in: awesome-ebpf · awesome-observability · awesome-opentelemetry · awesome-sre-tools · awesome-sre-agents · awesome-cloud-native · awesome-profiling · Awesome-GPU · awesome-gpu-engineering · awesome-helm · awesome-platform-engineering · awesome-k8s-tools · awesome-mcp-servers · awesome-devops-mcp-servers · MCP Registry · Glama · mcpservers.org
Version: 0.18.0
The only GPU observability tool your AI assistant can talk to.
"What caused the GPU stall?" → "forward() at train.py:142 - cudaMalloc spiking 48ms during CPU contention. 9,829 calls, 847 scheduler preemptions."
Ingero is a production-grade eBPF agent that traces the full chain - from Linux kernel events through CUDA API calls to your Python source lines - with <2% overhead, zero code changes, and one binary.
<img src="docs/assets/readme-demo-incident.gif" width="800" alt="ingero demo incident — CPU contention causes GPU latency spike, full causal chain diagnosis with root cause and fix recommendation">
docker run --rm --privileged --pid=host \ -v /sys/kernel/debug:/sys/kernel/debug \ -v /sys/kernel/btf:/sys/kernel/btf:ro \ -v /var/lib/ingero:/var/lib/ingero \ ghcr.io/ingero-io/ingero trace --record
Minimum capabilities (alternative to `--privileged`): `--cap-add=BPF --cap-add=PERFMON --cap-add=SYS_ADMIN`.
> **Note:** eBPF tracing (`trace`, `demo --gpu`) requires `--privileged --pid=host` plus the kernel volume mounts shown above. Without these, only unprivileged commands work (`demo --no-gpu`, `check`, `version`, `explain`, `query`). The `--pid=host` flag shares the host's `/proc` - do **not** also bind-mount `-v /proc:/proc:ro` as this causes OCI runtime errors on Docker Desktop and WSL2.
**Data persistence:** The container stores the SQLite database at `/var/lib/ingero/ingero.db` by default. Mount `-v /var/lib/ingero:/var/lib/ingero` to persist data after the container stops. Without this mount, **all trace data is lost** when the container exits.
**Multiple databases:** Use `--db` or the `INGERO_DB` env var to work with different databases:
bash
curl -fsSL https://raw.githubusercontent.com/ingero-io/ingero/main/scripts/install-deps.sh | bash
git clone https://github.com/ingero-io/ingero.git cd ingero make # generates eBPF bindings, builds, tests, and lints - single command sudo make install # optional - copies binary to /usr/local/bin/ingero # or just use ./bin/ingero directly, or: alias ingero=$PWD/bin/ingero ```
CONFIG_DEBUG_INFO_BTF=y)CAP_BPF + CAP_PERFMON (eBPF requires elevated privileges)VERSION=0.18.0 curl -fsSL "https://github.com/ingero-io/ingero/releases/download/v${VERSION}/ingero_${VERSION}_linux_amd64.tar.gz" | tar xz sudo mv ingero /usr/local/bin/
VERSION=0.18.0 curl -fsSL "https://github.com/ingero-io/ingero/releases/download/v${VERSION}/ingero_${VERSION}_linux_amd64.tar.gz" | tar xz sudo mv ingero /usr/local/bin/
VERSION=0.18.0 curl -fsSL "https://github.com/ingero-io/ingero/releases/download/v${VERSION}/ingero_${VERSION}_linux_arm64.tar.gz" | tar xz sudo mv ingero /usr/local/bin/ ```
Multi-arch images (amd64 + arm64) are published to GHCR on every release:
```bash
docker pull ghcr.io/ingero-io/ingero:v0.18.0
docker build -f deploy/docker/Dockerfile \ --build-arg VERSION=0.18.0 \ --build-arg COMMIT=$(git rev-parse --short HEAD) \ --build-arg BUILD_DATE=$(date -u +%Y-%m-%dT%H:%M:%SZ) \ -t ingero:local . ```
GHCR images have version info baked in automatically via GoReleaser. See deploy/docker/Dockerfile for details.
```bash
| Topic | Link |
|---|---|
| Full command reference (all 10 commands, every flag, output samples) | [docs/commands.md](docs/commands.md) |
| Architecture (single-node + cluster modes, ASCII diagram, pipeline stages) | [docs/architecture.md](docs/architecture.md) |
| Multi-node cluster deployment (K8s / bare-metal / Docker) | [docs/quickstart_fleet.md](docs/quickstart_fleet.md) |
| Stack tracing deep-dive (Python walker selection, JSON output, ptrace troubleshooting) | [docs/stack_tracing.md](docs/stack_tracing.md) |
| OTLP / Prometheus integration (transport, metric names, compatibility) | [docs/otlp.md](docs/otlp.md) |
| Full 25-problem detection catalog | [docs/detections.md](docs/detections.md) |
| AI investigation walkthrough (real session on A100 + GH200) | [docs/ml_eng_sample_investigation_session.md](docs/ml_eng_sample_investigation_session.md) |
Fleet ingero fleet-push subcommand | [docs/push_fleet.md](docs/push_fleet.md) |
| Multi-node remediation protocol (PoC, experimental) | [docs/remediation-protocol_fleet.md](docs/remediation-protocol_fleet.md) |
| Test matrix | [docs/test_matrix.md](docs/test_matrix.md) |
| Integration | What it does | Reference |
|---|---|---|
| **MCP server** (AI agents) | ingero mcp exposes 10 tools + a /investigate prompt to any MCP-compatible client (Claude Code, Cursor, Ollama). Stdio or HTTPS / TLS 1.3. | [docs/commands.md#ingero-mcp](docs/commands.md#ingero-mcp) |
| **OTLP / Prometheus** | --otlp HOST:PORT (HTTP JSON) or --prometheus :PORT (pull). Standard semantic conventions; compatible with OTel Collector, Grafana Alloy / Cloud, Datadog, New Relic. | [docs/otlp.md](docs/otlp.md) |
| **Browser dashboard** | ingero dashboard serves an HTTPS dashboard backed by the SQLite trace DB. Live ops / chains / snapshots. | [docs/commands.md#ingero-dashboard](docs/commands.md#ingero-dashboard) |
| **Multi-node cluster** | Real-time cluster-wide straggler detection via the Ingero Fleet collector. Agents push OTLP via ingero fleet-push. K8s / bare-metal / Docker. | [docs/quickstart_fleet.md](docs/quickstart_fleet.md) |
| **Perfetto / Chrome tracing** | ingero export --format perfetto produces a timeline you open in [ui.perfetto.dev](https://ui.perfetto.dev) or chrome://tracing. One track per node / rank. | [docs/commands.md#ingero-export](docs/commands.md#ingero-export) |
docker run --rm \ -v /var/lib/ingero:/var/lib/ingero \ ghcr.io/ingero-io/ingero query --db /var/lib/ingero/training-run-41.db --since 1h
docker run --rm \ -v /var/lib/ingero:/var/lib/ingero \ ghcr.io/ingero-io/ingero query --db /var/lib/ingero/training-run-42.db --since 1h
The image is ~10 MB (Alpine 3.20 + statically linked Go binary). When building the dev Dockerfile locally, pass version info via build args:
bash
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ Apache 2.0 — 宽松开源协议,可商用,需保留版权声明和 NOTICE 文件,含专利授权条款。
经综合评估,ingero 在MCP工具赛道中表现稳健,质量良好。如果你已有明确的使用需求,可以直接上手体验;如果还在评估阶段,建议对比同类工具后再做决策。
| 原始名称 | ingero |
| 原始描述 | 开源MCP工具:eBPF agent and MCP server for GPU causal observability。⭐82 · C |
| Topics | mcpcausal-tracingcudacuda-graphsebpfgpuc |
| GitHub | https://github.com/ingero-io/ingero |
| License | Apache-2.0 |
| 语言 | C |
收录时间:2026-05-22 · 更新时间:2026-05-22 · License:Apache-2.0 · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端