mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-12 14:19:00 +01:00
fix(chat): post a conversation turn's answer while an approval holds the task (#20179)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - People talk to agents in chat. The agent's final message becomes the chat reply. > - A chat turn can end while the task stays held. For example, the agent resubmits a hire approval and the approval is pending. > - The finalizer keeps the task in review for the approval (`governed_response_waiting`). That decision is not `conversation_turn_finished`. > - Run completion then treats the yielded turn as a control-plane wait. It records `presentationDecision: none` and posts no comment. > - The person never sees the agent's answer. The chat shows nothing after the request. > - This pull request counts a committed conversation turn as answered unless its own pending question or confirmation card presents it. > - The benefit is that the person always gets the reply, whatever state the task keeps. ## Linked Issues or Issue Description Refs #19971 (post the full final reply). That fix covers the reply text. This fix covers replies that were never posted. **What happened** In a chat, the person asked the CEO agent to change a proposed hire to a cheaper model and send it back for approval. The agent resubmitted the approval correctly. It called `paperclip_finish`, and Paperclip accepted the report with "task is still waiting for approval". The run succeeded and the task moved to `in_review`. The run event `run.presentation.resolved` recorded `commentAction: none` (reason `yielded_control_plane_wait` on master). The agent's final message never reached the chat. **Expected behavior** A chat turn's accepted final answer is always posted to the conversation. An approval, an execution stage or a dependency blocker can hold the task. It must not hide the reply. Only a question or confirmation card that the turn itself left pending presents the turn instead of its prose. **Steps to reproduce** 1. Open an Agent Chat with an agent. Link a hire approval in `revision_requested` to the chat task. 2. Ask the agent to change the hire and resubmit it. 3. The agent resubmits the approval and ends the turn with a yielded result and a response wake. 4. See that the chat shows no reply. The run is `succeeded` and its presentation decision is `none`. ## What Changed - `server/src/services/native-runtime/native-governance.ts`: move the pending-card query of `pendingNativeGovernance()` into `pendingNativeInteraction()`, with no change in behavior. Add `conversationTurnAnswered()`. It returns false only when a pending thread interaction (the same predicate as status arbitration) gates the turn. It checks cards on their own, so a pending execution stage does not hide a card. - `server/src/services/heartbeat/run-completion.ts`: a successful conversation run whose status decision committed now passes `conversationTurnFinished: true` to the presentation resolver, unless `conversationTurnAnswered()` is false. The existing `conversation_turn_finished` path is unchanged. - `server/src/services/heartbeat/run-completion.test.ts`: add a regression table with a real database. A pending approval, a revision-requested approval, a pending execution stage and a dependency blocker now post the answer. A finished turn still posts. The turn's own question card still owns the response, also when an execution stage is pending. - `doc/execution-semantics.md`: state the rule. Other finish outcomes I checked: - `done` or `needs_review` with a gate (`governed_gate_pending`): the result is not yielded, so the reply was already posted. - `blocked` (`paperclip_block`, waiting on a person or another task): not yielded, so the reply was already posted. - Yielded with a dependency blocker (`durable_dependency_blocker_bound`), or with `record_finalization_error` (`prior_status_preserved_no_live_path`, `monitor_wait_authority_lost`): the reply was also dropped. This fix covers them, because the rule does not depend on the reason code. ## Verification - `cd server && npx vitest run src/services/heartbeat/run-completion.test.ts`: 118 passed. Before the fix, 4 new cases failed (both approval states, execution stage, dependency blocker). The finished-turn and question-card cases passed both before and after. - `cd server && npx vitest run src/__tests__/heartbeat-run-summary.test.ts src/services/heartbeat/run-comment-selection.test.ts src/__tests__/slack-conversation-lifecycle.test.ts src/services/native-runtime/agent-chat-response-recovery.test.ts src/__tests__/chat-run-milestone-governed-work.test.ts src/__tests__/agent-conversations.test.ts`: all passed. - `tsc --noEmit -p server`: no errors in the changed files. Unrelated errors in this worktree come from an unbuilt local Runner package. - An end-to-end chat eval run on the case branch is in progress. I will add its result here. ## Risks - Low risk. The change only affects successful Agent Chat runs with a committed status decision. Ordinary tasks and external chat (Slack, GitHub) keep their own paths. External chat returns before this check. - A turn that yields while an approval is pending now posts its prose. The approval stays pending and the task status does not change. - One extra small read (the task row and pending gates) runs after a conversation turn that is not `conversation_turn_finished`. It runs after the provider turn ends, so it does not add time before the first token. - Previous-release state: no persisted shape changes. The fix reads `finalizationPhase` and `finalizationReasonCode`, which the previous release already writes. A run in flight at the deploy completes on the new code and posts its reply. A run that completed before the deploy keeps its recorded decision; nothing re-presents it. - Known gap, unchanged: background presentation recovery for conversations still requires `conversation_turn_finished` with a `done` result. If the server restarts between the status commit and the live presentation, a held yielded turn is not re-posted. That is a separate follow-up. > 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`), with tool use (shell, file edits, test runs) in Claude Code. ## 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 - [x] 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
1 parent
c5c587f162
commit
95bf90bd2c
4 files changed
+123
-19
No files matched your search
@@ -16,7 +16,7 @@ import { createHash, randomUUID } from "node:crypto";
|
||||
import { appendHeartbeatRunEvent } from "../heartbeat-run-events.js";
|
||||
import { createNativeRunLogCoverageVerifier, formatNativeRunLogEvent } from "../native-runtime/native-run-log-export.js";
|
||||
import type { Db } from "@paperclipai/db";
|
||||
import { activityLog, agents, agentWakeupRequests, budgetReservations, caseHandoffs, cases, companies, completionContracts, costEvents, createDb, EMBEDDED_POSTGRES_TEST_TIMEOUT_MS, heartbeatRuns, heartbeatRunEvents, issueComments, issueQuestionResponseDeliveries, issueRecoveryActions, issues, nativeRunFinalizations } from "@paperclipai/db";
|
||||
import { activityLog, agents, agentWakeupRequests, approvals, issueApprovals, budgetReservations, caseHandoffs, cases, companies, completionContracts, costEvents, createDb, EMBEDDED_POSTGRES_TEST_TIMEOUT_MS, heartbeatRuns, heartbeatRunEvents, issueComments, issueQuestionResponseDeliveries, issueRecoveryActions, issues, nativeRunFinalizations } from "@paperclipai/db";
|
||||
import { and, eq, sql } from "drizzle-orm";
|
||||
import { afterAll, beforeAll, beforeEach, describe, expect, it, onTestFinished, vi } from "vitest";
|
||||
import { getEmbeddedPostgresTestSupport, startEmbeddedPostgresTestDatabase } from "../../__tests__/helpers/embedded-postgres.js";
|
||||
@@ -469,6 +469,67 @@ describe.skipIf(!support.supported && !process.env.PAPERCLIP_TEST_DATABASE_URL)(
|
||||
expect(f.deps.finalizeAgentStatus).toHaveBeenCalledWith(f.agent.id, "succeeded", null, expect.any(Object));
|
||||
});
|
||||
|
||||
// A chat turn can finish while its task keeps waiting: a resubmitted hire
|
||||
// approval, an execution stage or a dependency blocker holds the status.
|
||||
// The accepted final answer must still reach the conversation. Only a card
|
||||
// the run itself left pending in the thread presents the turn instead.
|
||||
it.each([
|
||||
{ hold: "pending approval", reasonCode: "governed_response_waiting", replies: true },
|
||||
{ hold: "approval sent back for revision", reasonCode: "governed_response_waiting", replies: true },
|
||||
{ hold: "execution stage", reasonCode: "governed_response_waiting", replies: true },
|
||||
{ hold: "dependency blocker", reasonCode: "durable_dependency_blocker_bound", replies: true },
|
||||
{ hold: "finished turn", reasonCode: "conversation_turn_finished", replies: true },
|
||||
{ hold: "own question card", reasonCode: "governed_response_waiting", replies: false },
|
||||
{ hold: "execution stage and own question card", reasonCode: "governed_response_waiting", replies: false },
|
||||
] as const)("posts a waiting conversation turn's accepted answer: $hold", async ({ hold, reasonCode, replies }) => {
|
||||
const f = await fixture({ runtimeMode: "native" }, `CNV${randomUUID().slice(0, 6)}`);
|
||||
const [issue] = await db.insert(issues).values({ companyId: f.run.companyId, title: "Chat with CEO",
|
||||
status: "in_progress", assigneeAgentId: f.agent.id, responsibleUserId: "local-board",
|
||||
conversationAgentId: f.agent.id, conversationUserId: "local-board", conversationState: "active",
|
||||
...(hold.startsWith("execution stage") ? { executionState: { status: "pending" } } : {}),
|
||||
}).returning();
|
||||
await db.update(heartbeatRuns).set({ nativeIssueId: issue.id, responsibleUserId: "local-board",
|
||||
contextSnapshot: { issueId: issue.id, conversationSessionGeneration: 0 } }).where(eq(heartbeatRuns.id, f.run.id));
|
||||
if (hold.includes("approval")) {
|
||||
const [approval] = await db.insert(approvals).values({ companyId: f.run.companyId, type: "hire_agent",
|
||||
status: hold === "pending approval" ? "pending" : "revision_requested", payload: { name: "Researcher" } }).returning();
|
||||
await db.insert(issueApprovals).values({ companyId: f.run.companyId, issueId: issue.id, approvalId: approval.id });
|
||||
}
|
||||
if (hold.endsWith("own question card")) {
|
||||
await issueThreadInteractionService(db).create(issue, { kind: "ask_user_questions", sourceRunId: f.run.id,
|
||||
continuationPolicy: "wake_assignee", payload: { version: 1, questions: [{ id: "model", prompt: "Which model?",
|
||||
selectionMode: "single", required: true, options: [{ id: "flash", label: "Flash" }] }] } },
|
||||
{ agentId: f.agent.id, runId: f.run.id });
|
||||
}
|
||||
const answer = "Resubmitted the hire with the cheaper model. [Review the pending approval](/CNV/approvals/a1).";
|
||||
f.input.issueId = issue.id;
|
||||
f.input.issueRef = issue;
|
||||
f.input.issueContext = await createHeartbeatRunPreparation(db).getIssueExecutionContext(f.run.companyId, issue.id);
|
||||
// The finalizer's saved shape, unchanged from the previous release.
|
||||
f.input.adapterResult.resultJson = {
|
||||
nativeResult: { schema: "paperclip.run_result.v1", reportedWorkDisposition: "yielded",
|
||||
summary: "Waiting for the approval decision.",
|
||||
continuation: { kind: "response_wake", summary: "Wait for the decision.", idempotencyKey: "hire-wait" } },
|
||||
finalizationPhase: "committed", finalizationReasonCode: reasonCode,
|
||||
};
|
||||
f.deps.findLatestCompletedFinalAgentMessage.mockResolvedValue({ seq: 1, text: answer, sourceEventId: "final-event", channel: "final",
|
||||
reasonCode: "latest_non_empty_completed_final_agent_message" });
|
||||
f.deps.issuesSvc.addComment.mockImplementation(async (issueId, body) => {
|
||||
const [comment] = await db.insert(issueComments).values({ companyId: f.run.companyId, issueId,
|
||||
authorAgentId: f.agent.id, authorType: "agent", createdByRunId: f.run.id, body }).returning();
|
||||
return { ...comment, authorType: "agent" as const };
|
||||
});
|
||||
|
||||
await f.completion.completeRun(f.input);
|
||||
|
||||
const posted = (await db.select().from(issueComments).where(eq(issueComments.issueId, issue.id)))
|
||||
.filter(comment => comment.createdByRunId === f.run.id);
|
||||
expect(posted.map(comment => comment.body)).toEqual(replies ? [answer] : []);
|
||||
expect((await read(f.run.id)).resultJson).toMatchObject({ presentationDecision: replies
|
||||
? { chosenSource: "final_agent_message", commentAction: "create", commentId: posted[0]?.id }
|
||||
: { chosenSource: "none", commentAction: "none" } });
|
||||
}, EMBEDDED_POSTGRES_TEST_TIMEOUT_MS);
|
||||
|
||||
it("settles the accepted native exact rate-card price without replacing it with zero", async () => {
|
||||
const f = await fixture({ runtimeMode: "native" });
|
||||
const runnerInstanceId = randomUUID();
|
||||
|
||||
@@ -45,6 +45,7 @@ import { isUnresolvedWorkspaceBaseRefError, readUnresolvedWorkspaceBaseRefDiagno
|
||||
import { runnerGoalService } from "../runner-goals.js";
|
||||
import { resolveChatRunPresentationAuthorizationReason } from "../chat-run-publications.js";
|
||||
import { hasChatRunOwnedProviderInteraction } from "../chat-interaction-arbitration.js";
|
||||
import { conversationTurnAnswered } from "../native-runtime/native-governance.js";
|
||||
import { readAcceptedTaskContinuationSummary } from "../heartbeat-run-summary.js";
|
||||
import { SANDBOX_PROVIDER_PLUGIN_NOT_READY_REASON } from "../recovery/stranded-notice.js";
|
||||
import { MAX_TURN_CONTINUATION_RETRY_REASON, isResolvedInteractionContinuationWakeContext } from "../../modules/run-dispatch/index.js";
|
||||
@@ -845,8 +846,15 @@ export function createHeartbeatRunCompletion(db: Db, dependencies: HeartbeatRunC
|
||||
resultJson: persistedResultJson,
|
||||
interactionOwnsResponse: Boolean(issueId && readAcceptedTaskContinuationSummary(persistedResultJson) &&
|
||||
await hasChatRunOwnedProviderInteraction(db, { companyId: livenessRun.companyId, issueId, runId: livenessRun.id })),
|
||||
conversationTurnFinished: isConversation(issueContext) && livenessRun.status === "succeeded"
|
||||
&& persistedResultJson?.finalizationReasonCode === "conversation_turn_finished",
|
||||
conversationTurnFinished: Boolean(issueId) && isConversation(issueContext) && livenessRun.status === "succeeded"
|
||||
&& (persistedResultJson?.finalizationReasonCode === "conversation_turn_finished"
|
||||
// A turn left waiting on an approval, an execution stage or a
|
||||
// blocker still answered the person. Only its own question card
|
||||
// presents it instead.
|
||||
|| (persistedResultJson?.finalizationPhase === "committed"
|
||||
&& await measureRunPhase("response.resolve", () => conversationTurnAnswered({
|
||||
db, companyId: livenessRun.companyId, issueId: issueId!, runId: livenessRun.id,
|
||||
})))),
|
||||
existingComment: existingRunComment,
|
||||
finalAgentMessage,
|
||||
preferFinalResponseOverExistingComment:
|
||||
|
||||
@@ -24,6 +24,32 @@ export async function lockNativeGovernanceSources(input: { db: Db; companyId: st
|
||||
eq(approvals.companyId, input.companyId))).for("share", { noWait: true });
|
||||
}
|
||||
|
||||
/** The pending thread interaction (question or confirmation card) that gates this run, if any. */
|
||||
async function pendingNativeInteraction(input: {
|
||||
db: Db;
|
||||
companyId: string;
|
||||
issueId: string;
|
||||
runId: string;
|
||||
ignoredInteractionIds?: string[];
|
||||
}): Promise<{ id: string } | null> {
|
||||
const [issue] = await input.db.select({ conversationAgentId: issues.conversationAgentId, conversationUserId: issues.conversationUserId })
|
||||
.from(issues).where(and(eq(issues.id, input.issueId), eq(issues.companyId, input.companyId)));
|
||||
return input.db
|
||||
.select({ id: issueThreadInteractions.id })
|
||||
.from(issueThreadInteractions)
|
||||
.where(
|
||||
and(
|
||||
eq(issueThreadInteractions.companyId, input.companyId),
|
||||
eq(issueThreadInteractions.issueId, input.issueId),
|
||||
eq(issueThreadInteractions.status, "pending"),
|
||||
...(input.ignoredInteractionIds?.length ? [notInArray(issueThreadInteractions.id, input.ignoredInteractionIds)] : []),
|
||||
activeIssueInteractionCondition({ runId: input.runId, conversationMode: isConversation(issue) }),
|
||||
),
|
||||
)
|
||||
.limit(1)
|
||||
.then((rows) => rows[0] ?? null);
|
||||
}
|
||||
|
||||
export async function pendingNativeGovernance(input: {
|
||||
db: Db;
|
||||
companyId: string;
|
||||
@@ -37,23 +63,8 @@ export async function pendingNativeGovernance(input: {
|
||||
if (record(input.executionState).status === "pending") {
|
||||
return { kind: "execution_stage", id: input.runId };
|
||||
}
|
||||
const [issue] = await input.db.select({ conversationAgentId: issues.conversationAgentId, conversationUserId: issues.conversationUserId })
|
||||
.from(issues).where(and(eq(issues.id, input.issueId), eq(issues.companyId, input.companyId)));
|
||||
const [pendingInteraction, pendingApproval] = await Promise.all([
|
||||
input.db
|
||||
.select({ id: issueThreadInteractions.id })
|
||||
.from(issueThreadInteractions)
|
||||
.where(
|
||||
and(
|
||||
eq(issueThreadInteractions.companyId, input.companyId),
|
||||
eq(issueThreadInteractions.issueId, input.issueId),
|
||||
eq(issueThreadInteractions.status, "pending"),
|
||||
...(input.ignoredInteractionIds?.length ? [notInArray(issueThreadInteractions.id, input.ignoredInteractionIds)] : []),
|
||||
activeIssueInteractionCondition({ runId: input.runId, conversationMode: isConversation(issue) }),
|
||||
),
|
||||
)
|
||||
.limit(1)
|
||||
.then((rows) => rows[0] ?? null),
|
||||
pendingNativeInteraction(input),
|
||||
input.db
|
||||
.select({ id: approvals.id })
|
||||
.from(issueApprovals)
|
||||
@@ -75,3 +86,20 @@ export async function pendingNativeGovernance(input: {
|
||||
return null;
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* A committed conversation turn has answered the person, whatever status its
|
||||
* task keeps. An approval or an execution stage can hold the task, but neither
|
||||
* presents that answer in the conversation. Only a pending thread interaction
|
||||
* (a question or confirmation card) is the turn's visible response.
|
||||
*/
|
||||
export async function conversationTurnAnswered(input: {
|
||||
db: Db;
|
||||
companyId: string;
|
||||
issueId: string;
|
||||
runId: string;
|
||||
}): Promise<boolean> {
|
||||
// Check cards on their own: an execution stage outranks a card as the
|
||||
// status gate, but it does not present the turn.
|
||||
return !(await pendingNativeInteraction(input));
|
||||
}
|
||||
Reference in new issue
Block a user