Files
PaperClipAI/maps/agent-access-map/onboarding.md
T
ce27dde855 Start onboarding in chat and defer execution until the user chooses work (#22077)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work.
> - Onboarding connects an agent and gives the user a place to discuss
work.
> - The previous flow created an execution task and project before the
user chose a scope.
> - A standing chat can gather the goal and clear exit criteria first.
> - This pull request opens chat after onboarding and starts a live
welcome after the chat is visible.
> - The agent proposes work for acceptance, or creates a task when the
user explicitly requests one.
> - Subscription setup and shared question state must work for this
first conversation to succeed.

## Linked Issues or Issue Description

**Subsystem affected**
Company onboarding, agent chat, local subscription setup, task
numbering, and shared interaction views.

**Problem or motivation**
The first session creates work before the user has agreed to a goal.
Standing chats consume task numbers. An answered chat question can
remain pending in the inbox. Claude code entry requires an extra click,
and switching from Claude to Codex can leave an incompatible model
selected.

**Proposed solution**
Open an owner-bound chat for every new company. Start a live welcome
after the chat paints in a visible tab. Gather scope in conversation and
use the existing task acceptance flow. Keep chat records out of the task
number sequence. Publish question responses to shared caches and cancel
stale reads. Submit pasted Claude codes once and wait for verified
sign-in before connecting.

**Alternatives considered**
A fixed welcome would not use the onboarding context. A hidden execution
task would still create work and consume a task number. Keep the
existing task proposal and acceptance contract instead.

**Roadmap alignment**
This improves the shipped Agent Chat and Personal & Shared AI Accounts
flows in ROADMAP.md. It does not add a separate execution or approval
system.

Related prior work: Refs #19643 and #19512. Related route handoff work:
Refs #12557. This change replaces the automatic first task with chat; it
also repairs the subscription and shared question state paths used
during that journey.

## What Changed

- Finish onboarding in agent chat for default and team-template
companies. Create no execution task or project.
- Queue one live welcome after visible chat content paints. Pass company
goals and the initial idea as context. Preserve access, budget, and
hiring controls.
- Instruct the agent to gather a goal and exit criteria, offer an
interview, and propose tasks for acceptance. Allow direct creation on an
explicit user request.
- Leave new standing chats unnumbered. The first execution task receives
number 1. Preserve existing identifiers.
- Synchronize answered question cards across chat and inbox, including
concurrent requests and inactive cached views.
- Submit a pasted Claude authorization code once. Show connection
progress and hide the restart link while sign-in is still preparing.
Detect rejected OAuth codes promptly. Clear incompatible models when the
provider changes.
- Use `Work with [agent name]` as the default chat composer placeholder.
- Add regression coverage and update onboarding documentation and API
notes.

## Verification

- Completed a real Claude subscription journey on macOS with an isolated
local database. Authorized the account, pasted the real code, and
reached the review step without clicking Submit code or Connect. Get
started opened chat and the live welcome run succeeded. The company had
zero execution tasks, zero projects, and task counter 0.
- On rebased commit c664949858, 6 browser cases passed for
paste-to-connect-to-chat, task-free onboarding, existing-company
routing, the native connection probe gate, desktop/mobile planning, and
two-tab chat/inbox question synchronization.
- On the rebased head, 419 focused UI tests passed. The focused server
and adapter run passed 160 tests and hit one `socket hang up` in the
existing authenticated chat route test. The full 72-test conversation
suite passed on one repeat. The cause of that first socket failure
remains unconfirmed; the passing repeat does not establish its cause.
- UI typecheck and server declaration typecheck pass after the rebase.
The first server check found a missing local dependency from master; a
frozen-lockfile install repaired it. No dependency manifest or lockfile
change was needed.
- The first Greptile review found two welcome edge cases. Four new tests
reproduced both before the fix. On commit b380f14ef9, all 114 onboarding
and queue tests pass, including timeout retries and queued welcomes
replaced by a user message or chat reset. Server source and focused test
type checks pass on this revision.
- [Greptile
review](https://github.com/paperclipai/paperclip/pull/22077#issuecomment-6961533641)
gives rebased commit 2dbad34445 a score of
5/5. The current-head check passed. Both welcome review threads are
resolved, with no new actionable findings.
- CI serialized shard 8/9 found a missing OpenAPI inventory entry for
the new route file. The exact coverage test reproduced this locally.
Commit 17a72ec8b7 adds explicit mounted paths and documents both
board-only operations. All 65 OpenAPI and onboarding route tests and
server source typecheck pass after the fix. The final CI run passes on
this revision.
- Rebased onto master 23ce68a8dc and regenerated the SDK source registry
from committed inputs. Worktree preflight, SDK source validation, and
all 65 API/onboarding tests pass on final head
2dbad34445.
- On final head 2dbad34445, all 419
affected UI tests pass. UI package typecheck and server declaration
typecheck also pass. Commands: `pnpm --filter @paperclipai/ui
typecheck`; `node server/scripts/typecheck-layers.mjs
--declarations-only`; focused Vitest suites for the wizard, subscription
login, agent chat, shared interaction cache, and OpenAPI/onboarding
routes.
- [Final-head CI
run](https://github.com/paperclipai/paperclip/actions/runs/44208652564)
completed successfully on commit
2dbad34445. All required typecheck, build,
test, Runner, browser, serialized-server, and canary gates pass,
including the [final verify
job](https://github.com/paperclipai/paperclip/actions/runs/44208652564/job/134487749343).
The fresh routing gate selected the trusted AWS fleet after the
protected owner policy in #22098 merged. No source rebase was needed:
the gate reads that policy from protected master.
- The repaired OpenAPI shard passed in the earlier same-head CI run.
General server shard 8/12 hit a 20-second step timeout in the existing
`issue-recovery-live-run-agent-census` suite. This PR does not change
that suite or its fixture. All four tests pass locally on the exact
current commit in 84 seconds. The cause of the CI timeout remains
unconfirmed. Every other test job passed. A single failed-shard retry
was accepted but remained queued on its saved GitHub-hosted selection.
It was cancelled before test execution after #22098 merged. The fresh
full-CI run passed on the unchanged head under the approved AWS policy,
including [server shard
8/12](https://github.com/paperclipai/paperclip/actions/runs/44208652564/job/134481388730).
This passing execution does not establish the cause of the earlier
timeout.
- Worktree preflight, committed SDK source input validation, and diff
whitespace checks pass after the rebase.
- `pnpm verify` reports exit 2 because the new subsystem documentation
requires broader CI coverage. The final successful CI run supplies that
broader coverage. This is not a full local suite pass.

## Risks

- The initial task is now an explicit user decision. Existing onboarding
tasks and numbered records remain readable. No schema migration is
required.
- A welcome still depends on a working provider connection and the
normal run queue. Failed starts can be retried without duplicate
successful welcomes.
- Claude paste acceptance was verified with a real subscription on
macOS. This change does not claim a new live Codex or Windows acceptance
test.

## Model Used

OpenAI Codex, based on GPT-6, with tool-assisted reasoning, code
execution, terminal checks, and browser verification. The client does
not expose the exact routed build or context limit.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

---------

Co-authored-by: Scott Tong <scott@scottsmbpm5max.lan>
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-11-17 16:52:35 -08:00

16 KiB

Onboarding and first work — agent access

Product map: Onboarding and first work.

Feature ID: onboarding.

Review date: 2026-10-12. Source commit: f981fa3cc10cb8663a87dec60345fbff46d54d0a.

This is a source-reviewed baseline of representative operations and known gaps. Each capability needs operation-level review before its entire scope can be qualified. No live agent execution is claimed. Follow the map contract for format, permission checks, evidence and maintenance.

Scoped runner-default source review: 2026-10-12 at eee0534154eaf5d0d3f2bbc97c1e24d86f34a5a5, limited to first-agent selection, selected-runtime validation and draft restoration. Unrelated baseline entries retain the original source commit; no live agent verification is added.

instance

Capability ID: onboarding/instance.

Review: baseline.

API: not-applicable — Bootstrapping an instance cannot depend solely on an API served by that not-yet-running instance.

Agent access: operator-only — Existing CLI onboard/run/doctor / test-drive.

Policy: Instance setup is operator work; company creation is board; agent creation/task creation retain their individual permissions.

Evidence: source-only — implementation at the reviewed commit above.

Verification: not-run — exercise the listed local workflow in an authorized, owned operator environment. Verify its isolation and result; do not count local machine authority as company-agent API access.

Gaps: Bootstrapping an instance cannot depend solely on an API served by that not-yet-running instance.

company

Capability ID: onboarding/company.

Review: baseline.

API: present — representative operation IDs:

  • POST /api/companies.
  • POST /api/companies/{companyId}/agents.

Agent access: partial — Board company creation; authorized in-company agent hiring API.

Policy: Instance setup is operator work; company creation is board; agent creation/task creation retain their individual permissions.

Evidence: source-only — implementation at the reviewed commit above.

Verification: not-run — exercise a listed permitted operation with actual agent credentials/run context; inspect its result and applicable persistence/attribution. Assert the relevant denied actor, owner, company or mode from Policy. Review the remaining operations before qualifying the whole capability.

Gaps: API building blocks exist; an ordinary agent cannot create its parent company with a company key.

adapter

Capability ID: onboarding/adapter.

Review: baseline.

API: present — representative operation IDs:

  • GET /api/companies/{companyId}/adapters/{type}/models.
  • POST /api/companies/{companyId}/adapters/{type}/test-environment.

Agent access: partial — Permitted agent/model API; operator provider-account/environment setup.

Policy: Instance setup is operator work; company creation is board; agent creation/task creation retain their individual permissions.

The wizard selects the Codex harness and lets the server resolve an automatic new-agent runner without a runner dropdown. Explicit runner requests remain available through the agent creation API. Supported-platform and immutable sandbox evidence follow environment probe policy. Creation/discovery do not provision a target to determine this default. Adapter setup validation requires agents:create, plus saved-agent update access when testing an existing agent; interactive account login additionally requires a board user and owner-bound session. An agent's right to hire is not login or instance-environment administration authority.

First-agent onboarding limits the provider picker to Claude and Codex. Other providers remain available through normal agent creation; this UI choice does not change adapter APIs or hiring permissions.

A name-only, held setup has no provider selection yet, so the wizard omits its runner choice until the provider is chosen. The existing board-only setup and saved-agent update permissions remain; normal runner compatibility validation is unchanged. No runnable legacy fallback is created.

Scoped source-only runtime preparation: the existing authorized adapter test prepares the default Claude CLI in its already acquired sandbox when the image version is too old for the selected model. It installs a pinned package into a workspace-private versioned prefix and verifies the executable before the hello test. This adds no installation API, account grant, environment authority or engine fallback. Cloud UI omits manual shell debugging and legacy operator hints while retaining check messages and details. Local, SSH and custom commands remain operator-managed. Controls: runtime, probe, and wizard. Live managed authorization and subscription execution remain not-run for this change.

For fresh managed Codex Auto setup, the wizard's explicit Connect path can call POST /api/environments/{environmentId}/probe?companyId={companyId} before selected-runtime testing, then refresh scoped adapter discovery. This reuses the existing board instance-admin/local-implicit probe gate; agents:create alone is insufficient. A 403, absent proof, changed configuration, missing response identity or unconfirmed cleanup remains visible. No probe is added to creation or discovery, no agent/company grant is broadened, and provider authentication errors never cause an automatic runner switch.

Native setup derives outer sandbox isolation from the server-resolved execution target, as task execution does. This is an internal transport choice, not a request/configuration permission. Local and SSH setup keep Codex isolation; caller-supplied environment flags do not authorize an external sandbox. Existing company, account, probe and registration-cleanup gates still apply. The bounded hello probe retains read-only permissions and no dynamic tools. Its trusted launch skips project-instruction discovery in the new, empty setup workspace. This grants no filesystem or networking permission and does not change instruction discovery for ordinary task execution.

Scoped source review: 2026-10-13, exact baseline 1a0ca66242121e39d88455aa2b52a07acc4a07d9 plus the privately composed onboarding correction. Wizard controls and existing Product E2E observer controls exercise mocked allowed/denied setup and sticky unknown cost coverage. They do not prove live actor authorization or provider execution. The explicit manual confirmation-replies.runner-codex.daytona.task-card-accept case requires an eligible immutable OCI target and the existing probe authority, two identified setup leases and separately authorized paid/cleanup reservations. Its fixture neither prequalifies the environment nor creates the agent. Actual Cloud named snapshot binding, warm/cold signup and live allowed/denied agent checks remain not-run.

Evidence: source-only — implementation at the reviewed commit above.

Scoped Evidence: source-only — wizard/default selection, creation resolver and setup/login authorization at the scoped commit. Existing agent/approval execution preservation follows hiring access; no live onboarding proof is claimed here.

Verification: not-run — exercise a listed permitted operation with actual agent credentials/run context; inspect its result and applicable persistence/attribution. Assert the relevant denied actor, owner, company or mode from Policy. Review the remaining operations before qualifying the whole capability.

Gaps: Account authorization, environment administration and model selection need separate checks.

Verify automatic onboarding without a post-hire runner change, explicit Legacy, saved configuration after reload and the actual selected runtime's useful task plus follow-up. Assert denied agent login/probe and foreign-company account or environment access separately. These agent-context checks remain not-run.

first-task

Capability ID: onboarding/first-task.

Review: baseline.

API: present — representative operation IDs:

  • POST /api/companies/{companyId}/issues.
  • POST /api/issues/{issueId}/children.
  • POST /api/companies/{companyId}/chats/{agentRef}/onboarding (board only).
  • POST /api/companies/{companyId}/chats/{agentRef}/onboarding/greeting (board only).

Agent access: available — Native create_task / agent REST / CLI issue.

Policy: Instance setup is operator work; company creation is board; agent creation/task creation retain their individual permissions.

The company wizard opens the owner's standing chat without an execution task or project. Its board client requests a live welcome only once the thread is visible. Agents use the existing chat interactions to propose work; acceptance creates standalone tasks, and explicit task requests permit direct creation. See chat-first onboarding. The new board routes do not grant agents authority to start another person's onboarding conversation.

For historical task-based onboarding, an authenticated run bound to an onboarding first task must obtain persisted human acceptance of the latest ordinary proposal before creating new tasks through native create_task or REST. Its authenticated run determines this admission even for an explicit standalone task. A card decision or a saved conversational resolution with a genuine later same-task user message supplies acceptance; an agent-only resolution does not. A plan approval must name the current same-task revision. Newer pending/rejected proposals stop new creation. Already admitted task replay and ordinary delegation retain their existing rules. Native proposal metadata version and supersedeOnUserComment stays inside request_human_input.payload. Invalid root metadata is rejected before dispatch; repair guidance keeps the approval kind instead of replacing it with questions. The shipped recipes are checked against the advertised schema and the real interaction service. Idempotent proposal replay still needs a saved human decision. An accepted onboarding execution cannot create a new task blocked by its authenticated onboarding source. Native and REST creation return a recoverable input error before committing that task or its wake. The same source restriction applies to later blocker edits, using database UUID identity. A child's own agent run also cannot add its server-owned onboarding parent as a blocker. Native dependency edits carry their authenticated run. Empty-list corrections and other blockers remain available without a new proposal decision. The API keeps its normal attempt audit. A corrected creation call retains the human acceptance and can reuse the unconsumed command key. Other blocker relationships remain available. The child needs self-contained approved inputs; the parent uses set_dependencies to wait on the child, not the reverse.

Native paperclip_finish does not grant completion authority from an opening answer. The onboarding parent requires a trusted saved human decision on its latest proposal. Acceptance permits completion under the existing policy; rejection permits acknowledgment and closure without authorizing new work. The native status commit checks this decision again under its existing task/run locks. This adds no new agent endpoint or cross-company authority. The native completion controls check persisted outcomes on disposable PostgreSQL; they are not live Cloud credential or onboarding qualification. Source: creation admission. Controls: native/REST PostgreSQL regressions. These controls do not qualify a live Cloud provider journey.

Evidence: source-only — implementation at the reviewed commit above.

Verification: not-run — exercise a listed permitted operation with actual agent credentials/run context; inspect its result and applicable persistence/attribution. Assert the relevant denied actor, owner, company or mode from Policy. Review the remaining operations before qualifying the whole capability.

Gaps: Task creation is agent-accessible where permitted; scheduling/readiness does not follow from successful creation. The Codex first-task fixture's operator cost guard uses the existing board agent-configuration PATCH before admitting work. This is not evidence of agent permission to update its runtime; the listed agent-context checks remain not-run.

Scoped source review: 2026-10-13, the first-task asset correction against ea6c8eb5e16d61111968f633a66748b397b64a3e. The existing first-task instructions send file-producing work through the child's own execution and existing dependency/block controls. Native register_deliverable retains its current-task binding; this adds no cross-task file-registration authority or permission. Document-only work may use the existing child-document API inline. Acceptance of an execution proposal authorizes its selected tasks and proposed team within existing hiring, budget, and connection controls. An explicitly plan-only approval does not authorize implementation; requested separate approvals remain in force. The existing asset tests check wording only. The literal baseline live attempt failed child-file delivery and completion reporting; revised live authorization and execution remain not-run.

resume

Capability ID: onboarding/resume.

Review: baseline.

API: not-applicable — No server onboarding-draft API for this browser-local state; classify explicitly or intentionally make it a cross-client product feature.

Agent access: not-applicable — Browser localStorage wizard draft; persisted resources use their domain APIs.

Policy: Instance setup is operator work; company creation is board; agent creation/task creation retain their individual permissions.

Evidence: source-only — implementation at the reviewed commit above.

Verification: not-run — use the product recipe for the local/developer surface. Verify listed shared-resource reads separately; local presentation alone is not agent API evidence.

Gaps: No server onboarding-draft API for this browser-local state; classify explicitly or intentionally make it a cross-client product feature.

Scoped source review of draft restoration preserves harness/runner intent and invalidates obsolete setup-test results after a selection change. Browser draft continuity is not API or live-agent evidence, and opening a saved Legacy agent does not apply new-agent defaults.

The initial matching route opening preserves the authorized saved wizard step on bare and company-prefixed reloads. Explicit new-company or changed-company openings retain their navigation override and clear foreign company resource IDs. Existing component controls exercise these presentation and account boundaries. This scoped correction adds no server draft API, agent grant or proof of live actor authorization. The managed first-task eval still reloads before Get started and remains a separate live acceptance gate. The existing mocked-login browser case also checks the same company-scoped active session and saved selection through route reopening and a real reload, with one login start and no pending-login hire. These fixture assertions do not prove server authorization or live credential use.