# jamjet-labs/jamjet [Health: Active] **Category:** πŸ› οΈ Other Tools and Integrations **Repository:** https://github.com/jamjet-labs/jamjet **GitHub Stars:** 21 **npm Downloads (last month):** 133 **Views:** 0 **Installs:** 0 **Upvotes:** 0 **Directory Page:** https://allmcps.com/mcp/jamjet-labs-jamjet ## Description Durable, agent-native AI runtime with native MCP client + server and A2A support. Rust core for performance, Python authoring for ergonomics. Features graph-based workflows, durable execution, and multi-agent coordination. ## Claude Desktop Quick Installation Install path detected from listing signals. Uses `npx` (confidence: high): ```json "mcpServers": { "jamjet": { "command": "npx", "args": ["-y","@jamjet/mcp-shim"] } } ``` ## Documentation & README

⚑ JamJet

**The action-control plane for AI agents.** One policy file. One audit trail. Across hooks, guardrails, MCP gateways, SDKs, and custom runtimes. [![jamjet MCP server](https://glama.ai/mcp/servers/jamjet-labs/jamjet/badges/score.svg)](https://glama.ai/mcp/servers/jamjet-labs/jamjet) [![CI](https://img.shields.io/github/actions/workflow/status/jamjet-labs/jamjet/ci.yml?label=CI&style=flat-square)](https://github.com/jamjet-labs/jamjet/actions) [![PyPI](https://img.shields.io/pypi/v/jamjet?style=flat-square&color=f5c518)](https://pypi.org/project/jamjet) [![Maven Central](https://img.shields.io/maven-central/v/dev.jamjet/jamjet-runtime-core?style=flat-square&color=f5c518&label=maven)](https://central.sonatype.com/artifact/dev.jamjet/jamjet-runtime-core) [![crates.io](https://img.shields.io/crates/v/jamjet-policy?style=flat-square&color=f5c518&label=crates.io)](https://crates.io/crates/jamjet-policy) [![License](https://img.shields.io/badge/license-Apache%202.0-f5c518?style=flat-square)](LICENSE) [![Discord](https://img.shields.io/badge/Discord-join-5865F2?style=flat-square&logo=discord&logoColor=white)](https://discord.gg/SAYnEj86fr) [jamjet.dev](https://jamjet.dev) Β· [Quickstart](https://jamjet.dev/quickstart) Β· [Docs](https://jamjet.dev/concepts) Β· [Examples](https://jamjet.dev/examples) Β· [Blog](https://jamjet.dev/blog) Β· [Discord](https://discord.gg/SAYnEj86fr) [![Open in GitHub Codespaces](https://img.shields.io/badge/Open%20in-Codespaces-blue?style=flat-square&logo=github)](https://codespaces.new/jamjet-labs/jamjet?quickstart=1) [![Open in Gitpod](https://img.shields.io/badge/Open%20in-Gitpod-orange?style=flat-square&logo=gitpod)](https://gitpod.io/#https://github.com/jamjet-labs/jamjet)
--- > *Write the safety policy once. Run it everywhere your agents can act.* JamJet sits underneath your agent (Claude Code, OpenAI Agents SDK, MCP clients, LangChain, CrewAI, ADK, Spring AI, custom code) and enforces what prompts cannot: - πŸ›‘οΈ **Block unsafe tool calls** at runtime: database deletes, payments, file writes - βœ‹ **Pause for human approval** on risky actions, durably - πŸ’Έ **Cap cost** per agent, per run, per project - πŸ“’ **Record an audit trail** that survives a regulator's review - βͺ **Replay or resume** crashed runs from the last checkpoint **Keep your agent framework. Add JamJet where tool calls need control.** ![JamJet safety demo](https://raw.githubusercontent.com/jamjet-labs/jamjet/HEAD/demo.gif) ## See it in 60 seconds ```bash pip install jamjet jamjet demo unsafe-tool-call ``` No API key. No Docker. No cloud account. The model is mocked; the enforcement path is real. Three more demos run the same way: ```bash jamjet demo approval # pause-for-approval flow jamjet demo budget-cap # $0.05 cost cap jamjet demo mcp-tool-policy # MCP-shaped policy (preview of JamJet Gateway) ``` Works alongside **Claude Code Β· OpenAI Agents SDK Β· MCP clients Β· LangChain Β· CrewAI Β· ADK Β· Spring AI Β· LangChain4j**. ## On the JVM: one dependency Spring AI and LangChain4j teams get the same layer without writing glue code. The Spring Boot starter auto-injects JamJet advisors into your `ChatClient`: every call becomes durable and audited, with no changes to your application code. ```xml dev.jamjet jamjet-spring-boot-starter 0.1.0 ``` ```properties spring.jamjet.runtime-url=http://localhost:7700 spring.jamjet.approval.enabled=true # opt-in human-in-the-loop approval ``` Crash recovery, event sourcing, and a REST endpoint to approve or reject held actions. Falls back gracefully if the runtime is unreachable. Prefer no sidecar at all? **[JamJet Java Runtime](https://github.com/jamjet-labs/jamjet-runtime-java)** embeds durable execution directly in your JVM process: Java 21, no Docker, **8.9Γ— faster** than calling out to a REST sidecar ([benchmark and launch post](https://jamjet.dev/blog/zero-sidecar-durable-agents-java/)). Works with Spring AI, LangChain4j, and Google ADK. β†’ See a Spring Boot agent survive `kill -9` mid-run in **[`examples/loan-underwriter-agent`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/loan-underwriter-agent)**: resumes from disk checkpoints, gates disbursement on human approval, emits a signed receipt bundle. ## The same policy, everywhere Every agent toolchain is inventing its own safety layer. JamJet gives you one policy file and one audit trail across all of them. | Adapter | Install | Host | |---|---|---| | [`@jamjet/claude-code-hook`](https://npmjs.com/package/@jamjet/claude-code-hook) | `npm i -g @jamjet/claude-code-hook` | [Claude Code](https://docs.claude.com/en/docs/claude-code/hooks) PreToolUse hook | | [`@jamjet/mcp-shim`](https://npmjs.com/package/@jamjet/mcp-shim) | `npx -y @jamjet/mcp-shim ...` | Any [MCP](https://modelcontextprotocol.io) client (Claude Desktop, Cursor, …) | | [`@jamjet/openai-guardrail`](https://npmjs.com/package/@jamjet/openai-guardrail) | `npm i @jamjet/openai-guardrail` | [OpenAI Agents SDK](https://github.com/openai/openai-agents-js) tool guardrail (TS) | | [`jamjet.integrations.openai_guardrail`](https://pypi.org/project/jamjet/) | `pip install jamjet` | [OpenAI Agents SDK](https://github.com/openai/openai-agents-python) tool guardrail (Python) | | [`jamjet`](https://pypi.org/project/jamjet/) | `pip install jamjet` | Python SDK + runtime | | [`dev.jamjet:jamjet-spring-boot-starter`](https://central.sonatype.com/artifact/dev.jamjet/jamjet-spring-boot-starter) | Maven / Gradle | [Spring AI](https://spring.io/projects/spring-ai) ChatClient advisors | | [`@jamjet/cloud`](https://npmjs.com/package/@jamjet/cloud) | `npm i @jamjet/cloud` | TypeScript SDK + shared engine | | [`@jamjet/cli`](https://npmjs.com/package/@jamjet/cli) | `npm i -g @jamjet/cli` | Unified `jamjet audit show` / `jamjet approve` | All adapters load the same `policy.yaml`. All emit conformant audit JSONL to `~/.jamjet/audit/`. Run `jamjet audit show` to tail every decision across every adapter in one chronological view. JamJet does not replace the hook points these platforms give you. It makes them do more: Claude Code's PreToolUse hook gets a real policy engine, approval flow, and audit trail, and the same rules carry unchanged to OpenAI Agents SDK, MCP clients, Spring AI, and your own Python or TypeScript code. ### One policy, every adapter ```yaml # ~/.jamjet/policy.yaml version: 1 rules: - { match: "*delete*", action: block } - { match: "payments.*", action: require_approval } - { match: "shell.exec", action: block } ``` Drop this file in `~/.jamjet/`. Every adapter listed above uses it automatically. > Prompts are not a security boundary. The runtime is. β†’ Read **[When AI Deletes the Database](https://jamjet.dev/blog/when-ai-deletes-the-database/)** for why this is a runtime architecture problem, not a model problem. β†’ See the deeper **[durability demo at jamjet.dev/demo](https://jamjet.dev/demo)** for what happens when an agent crashes mid-tool-call. ## Policy in your own code Drop a policy beside your agent code. The runtime intercepts any matching tool call *before* it leaves the agent's process: `blocked_tools` are refused outright, `require_approval_for` pauses execution durably and waits for an out-of-band decision. Crashes don't lose the approval; execution resumes when it arrives. ```yaml # workflow.yaml policy: blocked_tools: - "*delete*" - "payments.refund" require_approval_for: - "database.*" - "payment.transfer" - "user.suspend" ``` **Python, with the hosted control plane:** ```python import jamjet jamjet.cloud.configure(api_key="jj_...", project="my-agent") jamjet.cloud.policy("block", "*delete*") jamjet.cloud.policy("require_approval", "database.*") # Every OpenAI / Anthropic call in this process is now policy-gated. ``` β†’ Runnable approval workflow in **[`examples/hitl-approval`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/hitl-approval)** Β· [Cloud Quickstart](https://docs.jamjet.dev/en/docs/cloud-quickstart) ## Where JamJet sits ```text Your Agent / Framework (LangChain Β· CrewAI Β· ADK Β· custom Β· MCP client) β”‚ β–Ό β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β” β”‚ JamJet Safety Layer β”‚ β”‚ policy Β· approval Β· budget Β· audit Β· replay β”‚ β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜ β”‚ β–Ό Tools Β· MCP servers Β· APIs Β· DBs Β· Agents ``` ## Use JamJet when your agent can… - call MCP servers or arbitrary tools - write to a database - send emails or Slack messages - trigger payments or external API calls - access customer data or PII - run for minutes/hours and needs to survive crashes - spend real model budget at scale - delegate to other agents ## What JamJet adds | Without JamJet | With JamJet | |---|---| | Agent crashes lose progress | Resume from the last checkpoint | | Tool calls rely on scattered app logic | Runtime policy blocks unsafe actions | | Human approval is custom glue | Approval is a durable workflow step | | Costs are discovered after the bill | Budgets enforced per agent / per run | | Audit evidence is stitched from logs | Append-only event log, signed export | | Memory is framework-specific | Pair with [Engram](#engram) for portable memory (MCP Β· REST Β· Python Β· Java) | | Frameworks stay siloed | MCP + A2A connect tools and agents | ## A full runtime underneath The safety layer runs on a durable, event-sourced execution engine. When you want more than enforcement, it's already there: - **Durable execution.** Event log, snapshots, crash recovery, deterministic replay. `agent.run_durable(...)` in [`examples/react-agent-durable`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/react-agent-durable). - **Sessions and memory.** Persistent `Session` threads across runs and restarts, with a governed Engram retrieve/record loop. [`examples/session-memory`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/session-memory). - **Multi-agent.** `Sequential`, `Parallel`, coordinator `Team`, and `Loop`; each sub-agent runs as its own governed durable execution. [`examples/team-multi-agent`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/team-multi-agent). Declare whole fleets in YAML with cron schedules: [`examples/fleet`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/fleet). - **Evaluation.** Batch eval with LLM-judge scoring ([`examples/eval-harness`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/eval-harness)) and trajectory regression diffs ([`examples/trajectory-eval`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/trajectory-eval)). - **Deploy.** `agent.deploy(runtime="local" | "self-host" | "cloud")` ships the same compiled IR to any runtime. [`examples/deploy-an-agent`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/deploy-an-agent). - **Dev loop.** `jamjet create` scaffolds a project; `jamjet dev` runs the whole local stack with one command. - **Web Companion.** A UI embedded in the runtime binary: graph view, timeline, state inspector, replay and fork controls. ## Works with your stack, not a replacement JamJet does not replace LangChain, LangGraph, CrewAI, Google ADK, Spring AI, or your custom agent code. Use those to build agent behavior. Use JamJet to control what happens at runtime. | You're using | Keep it for | JamJet adds | |---|---|---| | **LangChain Β· LangGraph Β· CrewAI Β· Google ADK Β· AutoGen** | Authoring agent behavior | Runtime safety: policy, audit, replay, approvals | | **LangSmith Β· Arize Β· Weights & Biases** | Observability and evaluation | Active enforcement (block at runtime) + durable recovery | | **Temporal Β· Orkes Β· DBOS** | General durable workflows | Agent-native primitives: policy on tool calls, MCP/A2A, memory | | **Google Β· AWS Β· Azure agent platforms** | Cloud-native ecosystems | Open-source, cloud-neutral governance that works on-prem | Want to build the official integration for *your* framework? **[Claim a slot](https://github.com/jamjet-labs/jamjet-examples/issues?q=is%3Aissue+is%3Aopen+label%3Awanted-integration)**: 8 slots open, the first 10 merged contributors get JamJet swag. ## Examples | Example | What it shows | |---------|--------------| | [`01-block-unsafe-tool`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/01-block-unsafe-tool) | A destructive tool call blocked before execution | | [`hitl-approval`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/hitl-approval) | Human approval as a first-class workflow primitive | | [`react-agent-durable`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/react-agent-durable) | A ReAct agent on the durable engine: event log, replay, park-on-429 | | [`team-multi-agent`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/team-multi-agent) | Multi-agent Teams, each sub-agent its own governed run | | [`loan-underwriter-agent`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/loan-underwriter-agent) | Spring Boot agent that survives `kill -9` and gates on human approval | | [`claims-processing`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/claims-processing) | Insurance pipeline: 4 specialist agents + HITL + audit | | [`eval-harness`](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/eval-harness) | Batch evaluation with LLM judge scoring | β†’ [All 33 examples](https://github.com/jamjet-labs/jamjet/blob/HEAD/examples/) ## Engram **[Engram](https://github.com/jamjet-labs/jamjet/blob/HEAD/runtime/engram-server/README.md)** is the JamJet ecosystem's memory layer for agents. Where JamJet provides durable *execution* (process can crash and resume), Engram provides durable *memory* (facts persist across runs and version cleanly via `supersede()`). Temporal knowledge graph, hybrid retrieval, conflict detection. Ships as a [Rust crate](https://crates.io/crates/jamjet-engram) (also bundled into the Rust runtime above), an [MCP server](https://registry.modelcontextprotocol.io/servers/io.github.jamjet-labs/engram-server) (Docker Β· GHCR), a [standalone Python library](https://pypi.org/project/jamjet-engram) ([github.com/jamjet-labs/engram](https://github.com/jamjet-labs/engram), 71% on LongMemEval-S), a [Python client](https://pypi.org/project/jamjet) for the MCP server, and a [Spring AI `ChatMemoryRepository`](https://central.sonatype.com/artifact/dev.jamjet/engram-spring-boot-starter). Comparison with Mem0/Zep β†’ [java-ai-memory.dev](https://java-ai-memory.dev). ## Architecture
Stack diagram ``` β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β” β”‚ Authoring Layer β”‚ β”‚ Python SDK | Java SDK | TypeScript SDK | YAML β”‚ β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€ β”‚ Compilation / Validation β”‚ β”‚ Graph IR | Schema | Policy lint β”‚ β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€ β”‚ Rust Runtime Core β”‚ Protocol Layer β”‚ β”‚ Scheduler | State SM β”‚ MCP Client | MCP Server β”‚ β”‚ Event log | Snapshots β”‚ A2A Client | A2A Server β”‚ β”‚ Workers | Timers β”‚ β”‚ β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”΄β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€ β”‚ Enterprise Services β”‚ β”‚ Policy | Audit | PII Redaction | OAuth | mTLS β”‚ β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€ β”‚ Runtime Services β”‚ β”‚ Model Adapters | Tool Execution | Engram Memory β”‚ β”œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€ β”‚ Storage β”‚ β”‚ Postgres (production) | SQLite (local) β”‚ β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜ ``` *"Engram Memory" here is the in-process distribution bundled with the Rust runtime. Engram also ships standalone; see [Engram](#engram).*
## Documentation Full docs at **[jamjet.dev](https://jamjet.dev/quickstart)** [Quickstart](https://jamjet.dev/quickstart) Β· [Concepts](https://jamjet.dev/concepts) Β· [Python SDK](https://jamjet.dev/python-sdk) Β· [Java SDK](https://jamjet.dev/java-sdk) Β· [YAML Workflows](https://jamjet.dev/yaml-workflows) Β· [REST API](https://jamjet.dev/api-reference) Β· [MCP](https://jamjet.dev/mcp) Β· [A2A](https://jamjet.dev/a2a) Β· [Eval](https://jamjet.dev/eval) Β· [Enterprise](https://jamjet.dev/enterprise) Β· [Observability](https://jamjet.dev/observability) Β· [CLI](https://jamjet.dev/cli) Β· [Deployment](https://jamjet.dev/deployment) ## Contributing Contributions welcome. See [CONTRIBUTING.md](https://github.com/jamjet-labs/jamjet/blob/HEAD/CONTRIBUTING.md). **Looking for a starter task?** - Build a [framework integration](https://github.com/jamjet-labs/jamjet-examples/issues?q=is%3Aissue+is%3Aopen+label%3Awanted-integration): 8 slots open, first 10 contributors get JamJet swag - Browse [good first issues](https://github.com/jamjet-labs/jamjet/labels/good%20first%20issue) - Join the conversation in [Discord](https://discord.gg/SAYnEj86fr) ## Community [GitHub Discussions](https://github.com/jamjet-labs/jamjet/discussions) Β· [Issues](https://github.com/jamjet-labs/jamjet/issues) Β· [Discord](https://discord.gg/SAYnEj86fr) ## License Apache 2.0. See [LICENSE](https://github.com/jamjet-labs/jamjet/blob/HEAD/LICENSE). ---
*Hosted control plane available at [app.jamjet.dev](https://app.jamjet.dev): traces, approval queue, audit retention, team projects. Optional. The runtime, all SDKs, and Engram are Apache-2.0 with no usage limits.* ### ⭐ Star JamJet if you believe agents need a runtime safety layer Built by Sunil Prakash · © 2026 JamJet Labs · jamjet.dev · Apache 2.0