feat(task-chat): approve harness-proposed plans (Codex plan mode, Claude ExitPlanMode) (#19291)

## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Agents run through the native runner. Codex and Claude Code each
have their own plan mode, and each produces a proposed-plan document.
> - Paperclip already has a planning work mode, an issue `plan`
document, and a `request_confirmation` approval gate for plans.
> - These parts were not connected. The Codex proposed-plan item was
transcript text only. The server path never put Codex into plan mode.
Claude was never put into plan mode, and its ExitPlanMode plan was
dropped. A planning run could close its task as done with no approval.
> - This pull request turns a harness's own proposed plan into the
issue's Plan document and the existing approval gate, for Codex and
Claude.
> - The benefit is that users approve or send back the agent's real plan
from the task chat. Approve resumes the agent. Request changes sends the
reason back to the agent.

## Linked Issues or Issue Description

Refs #19252

**Subsystem affected**
Native runner provider events (Codex and ACPX Claude), the runner Codex
provider (Rust), server native plan synchronization, and the task chat
UI.

**Problem or motivation**
Codex and Claude both produce a proposed-plan document in their native
plan mode. Paperclip did not use it. In live tests on the server path, a
Codex planning run was not in Codex plan mode, saved its own
`implementation-plan` document, and marked the task done without
approval. Claude planning runs were not in Claude plan mode, and the
ExitPlanMode plan text never reached Paperclip.

**Proposed solution**
Add a canonical `plan.proposed` event for a harness's proposed plan. Put
Codex and Claude into their native plan mode for planning runs. Save a
completed proposal as the issue `plan` document with the existing
`request_confirmation` gate. Agents can also propose a plan on a
standard task; the work mode stays standard.

**Alternatives considered**
Reuse `plan.updated`. Its schema is the within-turn checklist and is not
the durable Plan document, so a separate event keeps the two clear.
Stronger prompt text without native plan mode. Live tests showed that
the model kept its own document workflow.

**Roadmap alignment**
This reuses the issue plan document, planning work mode and approval
gate that already exist. It adds one canonical provider event.

## What Changed

- `packages/paperclip-runner/protocol/schemas/*`: add the
`plan.proposed` event (`paperclip.plan.proposed.v1`: `proposalId`,
`status`, `markdown` up to 60k characters, `source`). Regenerate the
schema bundle, validators and manifest.
- `packages/paperclip-runner/src/provider-events.ts`: map the Codex
`plan` item and a `<proposed_plan>` block in a completed Codex message
to `plan.proposed`. Add `claudePlanProposalEvent`.
-
`packages/paperclip-runner/runner/crates/runner-core/src/codex_provider.rs`:
send Codex's native `collaborationMode: plan` on `turn/start` for a
read-only (planning) Codex run. Before this change the server path never
sent it.
- `runner-core/src/acpx_event_payload.rs`: admit `plan.proposed` as ACPX
rich display activity.
- `drivers/acpx/codex-runtime-adapter.ts`: set Claude Code to native
plan mode for a read-only run. Capture Claude's ExitPlanMode plan from
its permission request, record it, and decline the native "Ready to
code?" prompt.
- `drivers/acpx/codex-acpx-driver.ts`, `cli/acpx-runtime-sidecar.ts`:
emit the captured proposal as a rich event. Complete that turn instead
of reporting it as cancelled.
- `backends/codex-native-backend.ts`: per-harness plan-mode
instructions, and a short opt-in hint for Claude on standard tasks.
- `server/src/services/native-runtime/native-session-executor.ts`:
synchronize a completed `plan.proposed` into the `plan` document and a
`request_confirmation` in planning and standard mode. Do not change the
issue status during the run (that revoked the run's lease). The run
finalizer turns the pending confirmation into the review wait.
- `server/src/services/heartbeat-run-summary.ts`: keep a
`<proposed_plan>` block out of the posted reply.
- UI: new `plan_proposal` provider family. The task chat shows a live
"Writing plan" card while a proposal streams and a one-line record when
it completes. Resolved plan reviews keep a receipt ("Approved Plan v1",
"Requested changes to Plan v1: reason"). Plan previews skip section
headings.
- Storybook: `Tasks and conversations/.../Task chat/Proposed plan
approval` shows the journey.
- `doc/SPEC-implementation.md`: describe harness-proposed plan approval.

## Verification

- `cargo test -p paperclip-runner-core --lib` (new:
`read_only_codex_turns_use_native_plan_collaboration_mode`)
- `cd packages/paperclip-runner && npx vitest run
src/provider-events.test.ts
src/drivers/acpx/codex-runtime-adapter.test.ts`
- `cd server && npx vitest run
src/__tests__/native-plan-proposal-sync.test.ts
src/__tests__/heartbeat-run-summary.test.ts`
- `cd ui && npx vitest run
src/components/transcript/native-run-events.test.ts
src/components/task-chat`
- Live test on a local worktree instance with real harnesses:
- Codex (`gpt-5.6-sol`), planning task: Codex asked one question, then
produced a native plan item. The task showed Plan v1 with Approve plan /
Request changes and moved to in review. Approve changed the work mode to
standard. Codex then implemented the plan and its tests passed.
- Codex, Request changes with a reason: Codex woke in planning mode and
produced Plan v2 that includes the change.
- Claude (`claude-opus-5-5`), planning task: Claude was in plan mode.
ExitPlanMode became Plan v1 and the turn completed. Approve resumed
Claude in standard mode, and Claude implemented the plan.
- Claude, standard task that asks for a plan first: Claude stopped
before code changes and asked for plan approval. The work mode stayed
standard.
- Storybook: open `Proposed plan approval` and check each stage.

## Risks

- Codex planning runs now really use Codex plan mode on the server path.
Codex can ask a clarifying question before it proposes. This matches
Codex plan mode.
- Claude planning runs now use Claude plan mode. A declined ExitPlanMode
ends the Claude turn; the runner reports it as completed. Claude's tool
row for that request records a rejection from Claude's side.
- Codex does not stream the proposed plan on the server path yet, so the
live card shows a placeholder until the plan completes.
- A `<proposed_plan>` block in any completed Codex message on a standard
task now asks for approval. This is the documented opt-in path.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Anthropic Claude Opus 5.5 (`claude-opus-5-5`), via Claude Code with
tool use (shell, file edits, browser). Long context. No extended
thinking mode was used.

## 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
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [ ] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
DottaandClaude Opus 5.5 authored and GitHub committed 2026-11-05 20:53:33 +00:00
1 parent 1b0f78100d
commit 22281d4b3a
51 files changed
+2015 -116

No files matched your search

@@ -1593,6 +1593,35 @@ export function continuingPendingInteractionIds(
.map((response) => response.interactionId);
}
/**
* A planning run can synchronize its completed checklist before the harness's
* full proposal arrives. The proposal then supersedes that same-run revision
* (its pending review expires as stale); a revision written by anyone else
* since the pinned base still fails the revision check.
*/
async function proposalBaseRevisionId(input: {
db: Db;
issueId: string;
runId: string;
pinnedRevisionId: string | null;
proposal: boolean;
}): Promise<string | null> {
if (!input.proposal) return input.pinnedRevisionId;
const latest = await documentService(input.db).getIssueDocumentByKey(input.issueId, "plan");
if (!latest?.latestRevisionId || latest.latestRevisionId === input.pinnedRevisionId) {
return input.pinnedRevisionId;
}
const latestRevision = await input.db
.select({ createdByRunId: documentRevisions.createdByRunId })
.from(documentRevisions)
.where(eq(documentRevisions.id, latest.latestRevisionId))
.limit(1)
.then((rows) => rows[0] ?? null);
return latestRevision?.createdByRunId === input.runId
? latest.latestRevisionId
: input.pinnedRevisionId;
}
export async function synchronizeCompletedProviderPlan(input: {
db: Db;
execution: NativeExecutionInput;
@@ -1603,25 +1632,55 @@ export async function synchronizeCompletedProviderPlan(input: {
payload: Record<string, unknown>;
};
}): Promise<PlanSynchronization | null> {
if (
input.event.eventType !== "plan.updated" ||
input.event.payload.complete !== true
)
return null;
if (
!("executionMode" in input.execution) ||
input.execution.executionMode !== "plan"
)
return null;
const planningContext = input.execution.planningContext;
if (!planningContext) return null;
const planId =
typeof input.event.payload.planId === "string"
const planMode =
"executionMode" in input.execution &&
input.execution.executionMode === "plan";
// A harness's own proposed-plan document (Codex plan item, Claude
// ExitPlanMode) asks for approval in any work mode: in Planning it is the
// planning deliverable, and in Standard the agent opted to stop and ask.
// A within-turn checklist only stands in for the plan in Planning mode.
const proposal =
input.event.eventType === "plan.proposed" &&
input.event.payload.status === "completed";
const completedChecklist =
input.event.eventType === "plan.updated" &&
input.event.payload.complete === true;
if (!proposal && !(completedChecklist && planMode)) return null;
// Only plan-capable work modes may write the Plan document. Ask and skill
// test runs answer in the thread; a proposal there must not become a plan.
const workMode = input.execution.task.workMode;
if (proposal && workMode !== "standard" && workMode !== "planning") return null;
const planningContext =
"executionMode" in input.execution && input.execution.executionMode === "plan"
? input.execution.planningContext
: null;
if (planMode && !planningContext) return null;
const baseRevisionId = planningContext
? await proposalBaseRevisionId({
db: input.db,
issueId: input.execution.binding.issueId,
runId: input.execution.binding.runId,
pinnedRevisionId: planningContext.baseRevisionId,
proposal,
})
: ((
await documentService(input.db).getIssueDocumentByKey(
input.execution.binding.issueId,
"plan",
)
)?.latestRevisionId ?? null);
const planId = proposal
? typeof input.event.payload.proposalId === "string"
? input.event.payload.proposalId
: ""
: typeof input.event.payload.planId === "string"
? input.event.payload.planId
: "";
const providerRevision = Number.isSafeInteger(input.event.payload.revision)
? Number(input.event.payload.revision)
: 0;
const providerRevision = proposal
? 1
: Number.isSafeInteger(input.event.payload.revision)
? Number(input.event.payload.revision)
: 0;
const body = providerPlanMarkdown(input.event.payload);
const digest = createHash("sha256").update(body).digest("hex");
if (!planId || providerRevision < 1 || !body) {
@@ -1630,7 +1689,7 @@ export async function synchronizeCompletedProviderPlan(input: {
planId,
providerRevision,
status: "invalid",
baseRevisionId: planningContext.baseRevisionId,
baseRevisionId: baseRevisionId,
digest,
documentRevision: null,
currentRevisionId: null,
@@ -1693,7 +1752,7 @@ export async function synchronizeCompletedProviderPlan(input: {
title: "Plan",
format: "markdown",
body,
baseRevisionId: planningContext.baseRevisionId,
baseRevisionId: baseRevisionId,
changeSummary: provenance,
createdByAgentId: input.execution.binding.agentId,
createdByRunId: input.execution.binding.runId,
@@ -1714,7 +1773,7 @@ export async function synchronizeCompletedProviderPlan(input: {
planId,
providerRevision,
status: "conflict",
baseRevisionId: planningContext.baseRevisionId,
baseRevisionId: baseRevisionId,
digest,
documentRevision: latest?.latestRevisionNumber ?? null,
currentRevisionId: latest?.latestRevisionId ?? null,
@@ -1731,7 +1790,7 @@ export async function synchronizeCompletedProviderPlan(input: {
planId,
providerRevision,
status: "conflict",
baseRevisionId: planningContext.baseRevisionId,
baseRevisionId: baseRevisionId,
digest,
documentRevision: current?.latestRevisionNumber ?? null,
currentRevisionId: current?.latestRevisionId ?? null,
@@ -1739,7 +1798,6 @@ export async function synchronizeCompletedProviderPlan(input: {
};
}
let confirmationId: string;
let confirmationPending = false;
try {
const confirmation = await issueThreadInteractionService(input.db).create(
{
@@ -1779,62 +1837,29 @@ export async function synchronizeCompletedProviderPlan(input: {
},
);
confirmationId = confirmation.id;
confirmationPending = confirmation.status === "pending";
} catch {
return {
eventId: input.event.sourceEventId,
planId,
providerRevision,
status: "approval_failed",
baseRevisionId: planningContext.baseRevisionId,
baseRevisionId: baseRevisionId,
digest,
documentRevision: revision.revisionNumber,
currentRevisionId: revision.id,
confirmationId: null,
};
}
if (confirmationPending) {
try {
const currentIssue = await input.db
.select({ status: issues.status })
.from(issues)
.where(eq(issues.id, input.execution.binding.issueId))
.limit(1)
.then((rows) => rows[0] ?? null);
if (currentIssue && currentIssue.status !== "in_review") {
await issueService(input.db).update(input.execution.binding.issueId, {
status: "in_review",
actorAgentId: input.execution.binding.agentId,
});
}
} catch {
const settledIssue = await input.db
.select({ status: issues.status })
.from(issues)
.where(eq(issues.id, input.execution.binding.issueId))
.limit(1)
.then((rows) => rows[0] ?? null);
if (settledIssue?.status !== "in_review") {
return {
eventId: input.event.sourceEventId,
planId,
providerRevision,
status: "approval_failed",
baseRevisionId: planningContext.baseRevisionId,
digest,
documentRevision: revision.revisionNumber,
currentRevisionId: revision.id,
confirmationId,
};
}
}
}
// Synchronization runs while the native run still owns the task. Changing
// the task status here would revoke the run's execution lease; the pending
// confirmation is a governance gate that the run finalizer turns into the
// review wait.
return {
eventId: input.event.sourceEventId,
planId,
providerRevision,
status,
baseRevisionId: planningContext.baseRevisionId,
baseRevisionId: baseRevisionId,
digest,
documentRevision: revision.revisionNumber,
currentRevisionId: revision.id,