agentd-core 1.3.4

Minimal, MCP-native agent runtime as a library: the agentic loop, supervisor, workflows, and code-registered tools (the agentd engine)
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61

# agentd-core

[agentd](https://agentd.dev) as a library: the agentic loop, the supervisor that
owns it, durable workflows, the A2A surface, and the config system — everything
the `agentd` binary is, minus the binary.

Most people want the binary. Reach for this crate when you need agentd's loop
*inside* your own process — an embedded agent in a larger service, a custom
front end over the same runtime, or tools registered in Rust rather than behind
an MCP server.

```toml
[dependencies]
agentd-core = "2.0"
```

```rust,no_run
use agentd::tools;

// Register a tool the model can call, in-process — no MCP server needed.
tools::register("deploy.status", "Current deployment status", |_args| {
    Ok(serde_json::json!({"env": "prod", "healthy": true}))
});
```

See `examples/embedded-agent.rs` in the repository for a complete one.

## What you get

- **The ReAct loop** — think, call a tool, observe, repeat, until an answer or a
  budget, with a context window that compacts rather than truncates.
- **A supervisor holding no model.** Lifecycle, limits and the process tree live
  in a component that cannot be prompted; the reasoning runs in child processes
  it can always kill. Cancellation is `killpg`, not a dropped future.
- **Durable workflows** — a graph checkpointed before every effect, so a
  restarted process rebuilds in-flight runs from the store and continues.
- **MCP and A2A**, implemented by [`rmcp`](https://crates.io/crates/rmcp) and
  [`a2a-rs`](https://crates.io/crates/a2a-rs) over
  [`agentd-net`](https://crates.io/crates/agentd-net)'s credentialed transport.
- **A security posture that is checked, not documented** — tools tagged
  untrusted-input / sensitive / egress, with the lethal trifecta refused at
  startup; secrets by reference only, never inline, never logged.

## Features

Capability is a compile-time decision. `tls` is on by default; `a2a`, `workflow`,
`cron`, `metrics`, `otel`, `oauth`, `aauth`, `hot-reload`, `config-watch`, `cel`
and `exec` are opt-in, and a flag whose feature is absent exits `2` loudly rather
than being ignored. MSRV is 1.96.

With the `a2a` feature, building this crate **from crates.io** needs `cmake` and
a C++ compiler: `a2a-rs` depends on `connectrpc`, which requests `rustls` with
its default features and so selects the C `aws-lc-rs` provider for the whole
graph. A build from [the repository] does not — it carries a vendored
`connectrpc` with that corrected, which a published crate cannot do for its own
dependencies. Until the fix is upstream, use the repository (or the release
binaries) if you want a pure-Rust build.

[the repository]: https://github.com/agentd-dev/source-code

Full documentation: <https://agentd.dev/docs/>. Licensed under Apache-2.0.