programmatic subagents
Clarify runtime capabilities and observation details
Updated README to clarify the capabilities of the runtime and added details about observing events.
authored by
Laurynas Keturakis
and committed by
GitHub
7387a006
840d90a6
+3
-2
+3
-2
README.md
+3
-2
README.md
···
1
1
# mill
2
2
3
-
A runtime for executing TypeScript programs that spawn and coordinate AI agents. Write plain TS with `await` and `Promise.all` — mill handles the lifecycle, persistence, and observability.
3
+
A runtime for executing TypeScript programs that can spawn agents. Write plain TS with `await` and `Promise.all` — mill handles the lifecycle, persistence, and observability.
4
4
5
5
## What you can do
6
6
···
10
10
11
11
**Swap agent backends** — drivers are generic adapters. Ship with pi, Claude, and Codex drivers. Write your own by implementing a codec that parses process output into structured events.
12
12
13
-
**Observe everything** — structured NDJSON event log per run, real-time streaming via `mill watch`, and full session replay via `mill inspect --session`.
13
+
**Observe everything** — structured NDJSON event log per run, real-time streaming via `mill watch`, and full session replay via `mill inspect --session`. Use the built-in `mill watch` or build a TUI around it.
14
14
15
15
## Quick example
16
16
···
27
27
agent: "planner",
28
28
systemPrompt: "You turn findings into an execution-ready plan.",
29
29
prompt: `Create remediation steps from:\n\n${scan.text}`,
30
+
model: "anthropic/claude-opus-4.6",
30
31
});
31
32
32
33
console.log(plan.text);