Files
paseo/public-docs/sdk/agents.md
T
0009a7e717 feat(workspaces): add background visibility (#5039)
* feat(server): let clients create internal agents

The daemon's own branch-name and commit-message generators run as
internal agents: never persisted, listed, flagged, announced, or seen by
plugin lifecycle hooks. Nothing outside the daemon could ask for one,
because the wire config schema strips `internal` and the agent.create
transform re-pins it. A plugin that runs a one-turn helper per event
therefore left an archived session in History every time.

Add an optional `internal` flag to `create_agent_request` (inherited by
`agent.create.request`), copy it onto the config in the session's create
path, and skip provider session persistence for it like the generators
do. The client SDK forwards the flag on both creation paths and gates it
once on `server_info.features.internalAgents`.

Two daemon paths assumed every agent is public and had to follow:
`resolveAgentIdentifier` now accepts a live agent by exact id, since the
id set it matched against came from listings that exclude internal
agents and left `wait_for_finish` answering "Agent not found"; and
archiving an internal agent now closes it without writing the record
that `registry.applySnapshot` used to leave behind, so `autoArchive`
does not accumulate hidden files on disk.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix(server): keep internal agents whole through failure, archive, and wait

Review follow-ups for internal agents:

- A create that fails after the runtime is registered now closes the
  internal agent before rethrowing, so a hidden runtime cannot outlive
  the failed request. Public agents keep today's behaviour.
- Archiving an internal agent cascades to attached children like any
  parent. cascadeArchiveChildren takes the parent record instead of
  reading it back from storage, where an internal agent never is.
- The archived snapshot and last message of an internal agent stay
  readable by exact id for ten minutes, so create-then-waitForFinish
  returns the finished turn whether the wait lands before or after an
  auto-archive. Storage plays that role for public agents.
- The internal-agent e2e scenarios live in the existing
  create_agent_request lifecycle suite rather than a new file.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

* fix(server): cascade to live internal children and keep a failed internal create's worktree until it closes

Internal children never reach storage, so the parent archive cascade scans
live agents for them as well. A failed internal create now clears the
created id only after the runtime is archived, so worktree cleanup cannot
delete the directory under a runtime that would not close. Stored agent
payloads carry lastError, so a retained failed helper reports its error
through wait_for_finish instead of the generic message.

* feat(cli,mcp): expose internal agents and document --no-wait

paseo run gains --internal and create_agent gains internal, both forwarded
to the create request. --no-wait becomes the documented spelling for
returning immediately, matching paseo send; --background and --detach stay
as hidden aliases under COMPAT tags.

* test: follow the --no-wait rename and stub getRetiredInternalAgent

* feat: list internal agents on request and add internal workspaces

Internal agents stay hidden by default, but a caller can now ask for
them: `fetch_agents_request` takes `filter.includeInternal`, snapshots
carry `internal: true`, and `paseo ls --internal` shows them in an
INTERNAL column. The app never passes the flag.

Workspaces get the same flag. `workspace.create.request` with
`internal: true` persists a hidden workspace: listings, update
subscriptions, sync reads, empty-project counting, open-by-path, MCP
`list_workspaces`, and plugin workspace hooks all skip it unless the
caller opts in with `filter.includeInternal`. Every agent created inside
an internal workspace is internal and unpersisted; AgentManager applies
that from the workspace id before the plugin hook so creators cannot
opt out. Archive and shutdown enumerate internal agents so nothing
inside an internal workspace outlives it.

Surfaces: `paseo workspace create --internal`, `paseo workspace ls
--internal`, `paseo run --internal --new-workspace`, MCP
`create_workspace({ internal })`, SDK `workspaces.create({ internal })`
and `list({ filter: { includeInternal } })`. Clients gate on
`features.internalWorkspaces`.

Co-Authored-By: Claude <noreply@anthropic.com>

* feat(workspaces): add background visibility with inherited creation defaults

* fix(cli): resolve workspace callers on the selected daemon

* test(workspaces): keep caller integration coverage in server suite

* test(cli): exercise workspace callers through CLI integration harness

* test(mcp): remove obsolete execution flags from browser fixtures

---------

Co-authored-by: marvin-ambrozio <marvin@gustavo.eng.br>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Co-authored-by: Mohamed Boudra <boudra.moha@gmail.com>
2026-10-09 09:31:50 +01:00

6.9 KiB

title, description, nav, order, category
title description nav order category
Agents with the SDK Create, run, find, reuse, parent, and archive coding agents from TypeScript. Agents 52 TypeScript SDK

Agents with the SDK

An agent handle keeps a stable agent ID and exposes the turn lifecycle without exposing daemon RPCs.

Run an initial prompt

const agent = await client.agents.create({
  config: {
    provider: "claude/claude-sonnet-5",
  },
  cwd: "/Users/me/dev/storefront",
  prompt: "Review the checkout flow and propose one focused fix.",
  labels: { source: "checkout-review" },
});

const result = await agent.waitForFinish();
console.log(result.status, result.lastMessage);

waitForFinish() returns one of four statuses:

Status Meaning
idle The turn completed and the agent can accept another prompt.
permission The agent needs a person to answer a permission request in Paseo.
error The provider ended the turn with an error.
timeout The wait deadline elapsed; the agent may still be running.

A timeout does not cancel the agent.

Keep a session alive for follow-ups

Create an idle session when prompts arrive later:

const reviewer = await client.agents.create({
  config: {
    provider: "codex/gpt-5.5",
  },
  cwd: "/Users/me/dev/storefront",
  title: "Checkout reviewer",
});

const first = await reviewer.run("Review the current diff.");

if (first.status === "idle") {
  const second = await reviewer.run("Now focus on failure recovery.");
  console.log(second.lastMessage);
}

Use send() for fire-and-forget delivery. Use run() when the caller needs the outcome of that turn.

Find agents by label

Set labels at creation, then filter on them. The daemon does the matching:

const page = await client.agents.list({
  filter: { labels: { "issue-provider": "my-tracker" } },
});

for (const { agent } of page.entries) {
  console.log(agent.id, agent.title, agent.status);
}

Continue an agent by ID

const agent = client.agents.ref("agent_01H8X...");

const result = await agent.run("Now write the fix.");
console.log(result.lastMessage);

ref() does not contact the daemon. Call refresh() first when you need to know whether the agent still exists; it returns null if it does not.

Create a subagent

Create a child through its workspace. The handle owns placement, so the caller does not repeat its directory:

if (!parent.workspaceId) throw new Error("Parent has no workspace");

const workspace = client.workspaces.ref(parent.workspaceId);
const child = await workspace.agents.create({
  config: {
    provider: "codex/gpt-5.5",
  },
  parent,
  title: "Implement checkout fix",
  prompt: "Implement the accepted checkout plan and run focused tests.",
});

parent establishes parentage. Archiving a parent cascade-archives its children. Call detach() first when a child should continue independently.

Run work in a background workspace

Create a background workspace and its first agent in one call. Put background on the workspace options and the agent's configuration under agent:

const cwd = "/Users/me/dev/storefront";
const workspace = await client.workspaces.create({
  source: { kind: "directory", path: cwd },
  background: true,
  agent: {
    cwd,
    config: { provider: "codex/gpt-5.5" },
    prompt: "Review the checkout flow.",
  },
});

The call returns a workspace handle and starts its first agent with the supplied prompt.

Background agents keep their history and survive daemon restarts. Default discovery hides them; use client.agents.list({ filter: { includeBackground: true } }) to include them. Exact-ID handles work normally. See workspace creation for defaults and caller inheritance.

Request structured output

const schema = {
  type: "object",
  properties: {
    summary: { type: "string" },
    risk: { type: "string", enum: ["low", "medium", "high"] },
  },
  required: ["summary", "risk"],
  additionalProperties: false,
};

const agent = await client.agents.create({
  config: {
    provider: "codex/gpt-5.5",
  },
  cwd: "/Users/me/dev/storefront",
  outputSchema: schema,
  prompt: "Assess the release risk of the current diff.",
});

const result = await agent.waitForFinish();
if (result.status !== "idle" || !result.lastMessage) {
  throw new Error(result.error ?? "Agent returned no structured output");
}

const assessment = JSON.parse(result.lastMessage) as {
  summary: string;
  risk: "low" | "medium" | "high";
};

Validate the parsed value in your application before using it as trusted input.

Read the current state of an agent

const agent = client.agents.ref("agent_01H8X...");
await agent.refresh();

if (agent.pendingPermissions?.length) {
  console.log(`Waiting on ${agent.pendingPermissions.length} permission request(s)`);
}
console.log(agent.lastUsage?.totalCostUsd, agent.runtimeInfo?.sessionId);

The handle exposes status, capabilities, availableModes, pendingPermissions, activeTurn, lastUsage, lastError, features, runtimeInfo, archivedAt, workspaceId, and cwd as properties. All of them read the last snapshot the handle observed and never fetch.

A handle from ref() has observed nothing, so every one of them is null until refresh(), run(), waitForFinish(), a timeline refetch, or subscribe() delivers a snapshot. Optional values in an observed snapshot also read as null. Use current() when you need the whole snapshot or need to distinguish those states.

subscribe() is a local listener. An owned agent-directory observation supplies its updates:

const unsubscribe = agent.subscribe(() => {
  if (agent.status === "error") console.error(agent.lastError);
});
const directory = await client.agents.list({ subscribe: {} });

// When this view closes:
unsubscribe();
await directory.subscription.release();

List the commands a session loaded

const { commands, error } = await agent.commands();
if (error) throw new Error(error);

const skills = commands.filter((command) => command.kind === "skill");

The answer comes from the running session, not from a directory scan, so it includes commands and skills built into the provider that never appear on disk. kind is the provider's own classification and is optional; treat a missing kind as unclassified rather than assuming "command".

A provider that cannot produce a list reports that in error and returns an empty commands array. The call does not reject.

Archive or detach

await agent.archive(); // Soft-deletes the agent and closes its runtime.
await child.detach(); // Keeps the child alive but removes its parent relationship.

Closing the SDK connection does not archive agents. Archive temporary agents explicitly, preferably in finally.