mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-10-29 04:25:52 +01:00
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Users connect accounts and choose an agent harness and model. > - The runtime change in #14970 supports custom providers on those connections. > - Normal setup must stay simple while advanced users can choose a compatible gateway. > - Shared connector rows and access controls keep these choices consistent. > - This pull request refines the agent setup UI and adds review stories and repeatable browser qualification. > - The qualification checks real tools and downloaded outputs, not only a successful run status. ## Linked Issues or Issue Description Refs #14970, #37, #13083, #14104, #14565, #12692. The core implementation in #14970 is merged. This branch incorporates its squash commit and targets `master`. Both PRs contain our implementation. #14016 is a reference only and is not a dependency. This PR has 96 changed files. ## What Changed - Complete model-provider connector presentation beside other connectors. Each row uses the existing Connect action and connection list. Tags are stored without category UI. The base PR includes the provider forms and routes. - Show persistent Subscription, API Key, and Advanced choices. Label Advanced as Custom Gateway. Reuse provider logos, connection lists, and permissions controls. Default access to the organization and all agents when permitted; keep narrowing controls under Advanced. - Keep Configure reachable before subscription sign-in, so users can select a supported environment when the default cannot sign in. Testing and saving still require a connection. Show the execution environment in Configure. Preserve the confirmed Connect choice. Editing a method, credential, saved account, or advanced choice requires that current choice to connect before testing or saving. Use matching model and thinking-effort dropdowns and retain connection icons in selected values. - Preserve the new harness model default when switching an existing OpenCode agent to Codex or Claude, and resolve user-selected model names with the effective harness. - Load popular OpenRouter models through the shared connection-model discovery path. Keep explicit model lists and manual model entry available. - Group onboarding, connection setup, agent runtime, management, recovery, and production-component stories under AI Connections / Provider routing. - Add an explicit-only provider-connections browser suite for managed local or existing local/staging targets. Use private browser profiles and credential handoffs. Support human-assisted subscription sign-in without sharing passwords or tokens in reports. - Verify persisted connection identity, runtime probes, tool execution, exact artifact bytes, completion, and context-dependent follow-up. Retain source/model provenance, cost bounds, closed error diagnostics, original failures, and cleanup evidence. - Add Gemini startup-model and skill-root fixes, Grok private-history detection, ACP filesystem regression fixtures, selected-workspace handling for local Hermes, and artifact-helper workspace fallback. - Keep managed Grok runtime homes disposable. Remove host-side transcript retention/restoration because private file modes do not isolate same-user agent processes. Ignore earlier development archives and use a fresh task handoff when history is unavailable. Verify the absence of restored transcripts with a separate same-user process. - Capture stopped-run diagnostics before deleting an attached-company fixture agent. Track creation and owned sign-in receipts; revoke only this attempt's accounts and never adopt a concurrent campaign's newly created account. Preserve failure signals and final status through cleanup. - Require the requested environment in the saved agent and every run, including follow-ups. Reject a forced incompatible target. Keep one cancellation state through startup, every cell, reporting, and teardown for SIGINT, SIGTERM, and SIGHUP. Stop further paid cells after interruption. Document qualification limits. ## Verification - Current head `b3bb3e94d577d43d9965a6b9daba039f599b2e49` includes master `d9f600043`. The security fix in `a758fde31` passes full workspace typecheck, production build, and 119 connection/Grok regressions. The unchanged UI passes all 126 configuration/model-discovery tests and token gates. The final published-guide correction passes Grok adapter typecheck. Earlier head `eebd8225c` passed the complete deterministic runner suite (1,404 Vitest tests and 128 Node tests) and all CI jobs. Current-head CI run `37520147514` passed all 47 jobs, including the full sharded Vitest and browser matrix, production build, and canary dry run. All 55 checks completed: 53 successes and two expected skips. The current-head security scan passed, Greptile is 5/5, and no review threads remain open. - A separate same-user process reproduced reading a restored Grok transcript before the security fix. The regression now finds no transcript. Existing fresh-session fallback and ordinary session metadata behavior pass. - The final account-choice and cleanup fixes pass 85 setup tests and 26 qualification-harness tests. Regressions verify that editing a connection invalidates confirmation, Configure remains reachable before sign-in, diagnostics are captured before fixture deletion, and concurrent campaigns cannot adopt or revoke each other's accounts. UI and E2E typechecks pass. - The Storybook build and actual Chromium production-component stories passed during this change. Review the neighboring AI Connections / Provider routing stories, regular connector rows, three connection modes, model discovery, and the single execution-environment control in Configure. - Cancellation smoke verified authenticated cleanup before browser close for SIGINT, SIGTERM, and SIGHUP. Regressions cover interruption during startup and reporting, missing-file ACP resource errors, and preserved permission denials. Both ACP runtime versions and 54 ACPX/Grok regressions passed. The deterministic connection-intent browser suite passed two tests. - Historical local qualification retained 43 passing API/gateway cells out of 46, with downloaded outputs and follow-up receipts. These attempts span earlier builds; they do not qualify this exact commit or staging. Subscription combinations, Gemini overloads, and the unresolved follow-up failure remain recorded rather than counted as passing. - Use `pnpm test:e2e:runner -- --list --suite provider-connections` to inspect the matrix. Follow `tests/runner-e2e/PROVIDER-CONNECTIONS.md` for credentials, target URL, sign-in assistance, budget, evidence, and cleanup. Paid live tests remain opt-in. ## Risks - The core implementation in #14970 is merged. This PR adds no database migration of its own. - Subscription login needs an interactive provider session. Dedicated accounts and staging qualification remain follow-up work; this PR does not certify every login combination for production. - Managed Grok transcript resume is deferred until provider history has an OS isolation or authorized broker solution. Follow-ups start fresh with Paperclip task context; earlier live Grok results do not qualify this behavior. - Gemini CLI 0.58.0 has an upstream ACP new-file error conversion defect. Live overloads and one unresolved follow-up timeout remain recorded. The stock CLI is unchanged, and those cases are not marked as passing. - Real-provider tests spend credits and use private credential/evidence directories. The launcher requires explicit selection and checks target ownership. It must not attach to a developer's database by accident. - OpenClaw Gateway, Hermes Gateway, Claude Managed, AWS AgentCore, Process, HTTP, and legacy ACPX local remain outside custom provider setup. ## Model Used OpenAI GPT-6 through Codex, with reasoning, repository tools, code execution, and browser testing. The exact deployment model ID and context window size were not exposed in this session. ## 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: Paperclip <noreply@paperclip.ing>
242 lines
9.0 KiB
TypeScript
242 lines
9.0 KiB
TypeScript
import { useRef, type ReactNode } from "react";
|
|
import { AnimatePresence, motion } from "motion/react";
|
|
|
|
import { cn } from "../../lib/utils";
|
|
import {
|
|
SOURCE_EXIT_FADE,
|
|
SOURCE_COLLAPSE_MOVE,
|
|
TAG_SWAP_ENTER,
|
|
TAG_SWAP_EXIT,
|
|
TAG_SWAP_TRAVEL,
|
|
} from "./onboarding-motion";
|
|
|
|
/**
|
|
* The connect step's row of model sources, and the tag under each one saying
|
|
* which credential the source would be reached with.
|
|
*
|
|
* Presentational only — the caller owns which source is picked and which
|
|
* credential mode is in force, because both outlive this row: the mode is set
|
|
* by a checkbox that sits below it, and the selection drives the step's CTA.
|
|
*/
|
|
|
|
/** How a source gets authenticated. Every tile is in the same mode at once. */
|
|
export type CredentialMode = "subscription" | "api";
|
|
export type ModelConnectionMode = CredentialMode | "advanced";
|
|
|
|
export type ModelSource = {
|
|
id: string;
|
|
label: string;
|
|
/** The brand mark, rendered into a 30px square. */
|
|
icon: ReactNode;
|
|
/** Override the row's credential mode; null leaves the tag slot empty. */
|
|
credentialMode?: ModelConnectionMode | null;
|
|
};
|
|
|
|
const CREDENTIAL_TAG_LABEL: Record<ModelConnectionMode, string> = {
|
|
subscription: "Subscription",
|
|
api: "API key",
|
|
advanced: "Custom Gateway",
|
|
};
|
|
|
|
/**
|
|
* The credential tag, swapping in a fixed-height slot.
|
|
*
|
|
* The slot has to hold its height whatever is in it: the tag is the last line
|
|
* of the tile, and a label that measured itself would resize the tile mid-swap
|
|
* and nudge the two beside it. `overflow-hidden` is doing real work too — it is
|
|
* what makes the outgoing label fall out of frame rather than slide past the
|
|
* tile's padding and over the row below.
|
|
*/
|
|
export function CredentialTag({ mode }: { mode: ModelConnectionMode }) {
|
|
return (
|
|
<span className="relative flex h-4 w-full items-center justify-center overflow-hidden text-(length:--text-micro) text-muted-foreground">
|
|
<AnimatePresence initial={false} mode="sync">
|
|
<motion.span
|
|
key={mode}
|
|
className="absolute inset-0 flex items-center justify-center whitespace-nowrap"
|
|
initial={{ opacity: 0, y: TAG_SWAP_TRAVEL }}
|
|
animate={{ opacity: 1, y: 0, transition: TAG_SWAP_ENTER }}
|
|
exit={{ opacity: 0, y: TAG_SWAP_TRAVEL, transition: TAG_SWAP_EXIT }}
|
|
>
|
|
{CREDENTIAL_TAG_LABEL[mode]}
|
|
</motion.span>
|
|
</AnimatePresence>
|
|
</span>
|
|
);
|
|
}
|
|
|
|
function ModelSourceTile({
|
|
source,
|
|
mode,
|
|
selected,
|
|
onSelect,
|
|
buttonRef,
|
|
settling,
|
|
}: {
|
|
source: ModelSource;
|
|
mode: ModelConnectionMode | null;
|
|
selected: boolean;
|
|
onSelect: () => void;
|
|
buttonRef: (node: HTMLButtonElement | null) => void;
|
|
settling: boolean;
|
|
}) {
|
|
return (
|
|
<button
|
|
ref={buttonRef}
|
|
type="button"
|
|
role="radio"
|
|
aria-checked={selected}
|
|
onClick={onSelect}
|
|
className={cn(
|
|
"flex min-w-0 flex-1 cursor-pointer flex-col items-center gap-1.5 self-stretch rounded-md border p-3",
|
|
// Longer while the row is settling back to its default. Dropping the
|
|
// selection is the last thing that happens on the way out, and at the
|
|
// interaction duration it landed as a colour swap after everything else
|
|
// had stopped — a cut rather than a release. Across the tile's travel
|
|
// it reads as the choice being let go.
|
|
"transition-(--tp-border-color-background-color) ease-(--motion-ease-standard)",
|
|
settling ? "duration-(--motion-duration-slow)" : "duration-(--motion-duration-fast)",
|
|
// Focus is a ring, never a border. The stroke has exactly one job here
|
|
// and lending it to focus as well would mean tabbing across the row
|
|
// looked like picking every tile in turn.
|
|
"outline-none focus-visible:ring-ring/50 focus-visible:ring-(length:--rad-3)",
|
|
// Selection is a lighter surface *and* a brighter edge. An earlier pass
|
|
// here used the fill alone, reasoning that a bright stroke on one tile
|
|
// made the row read as one outlined object beside a plain one. The
|
|
// design does both, and it is right: at these sizes one step of fill is
|
|
// too quiet to answer "which did I pick?" from across the screen, and
|
|
// the stroke is what carries it.
|
|
//
|
|
// Hover stops short of the selected fill, so pointing at a tile says
|
|
// "this one is live" rather than "this one is chosen".
|
|
selected
|
|
? "border-foreground/40 bg-accent"
|
|
: "border-border bg-card hover:bg-accent/40",
|
|
)}
|
|
>
|
|
<span className="flex size-(--sz-30px) shrink-0 items-center justify-center">
|
|
{source.icon}
|
|
</span>
|
|
{/*
|
|
One step up the named ladder each — the source name from text-xs (12px)
|
|
to --text-compact (13px), the tag under it from --text-nano (10px) to
|
|
--text-micro (11px), keeping the two a step apart. Both use the
|
|
font-size-only token form, so the line box comes from the tile's own
|
|
rhythm rather than the Tailwind scale's paired line-height.
|
|
*/}
|
|
<span className="text-(length:--text-compact) font-medium text-foreground">
|
|
{source.label}
|
|
</span>
|
|
{mode ? <CredentialTag mode={mode} /> : <span aria-hidden className="h-4" />}
|
|
</button>
|
|
);
|
|
}
|
|
|
|
export function ModelSourceTiles({
|
|
sources,
|
|
mode,
|
|
selectedId,
|
|
onSelect,
|
|
label,
|
|
collapsed = false,
|
|
settling = false,
|
|
}: {
|
|
sources: ModelSource[];
|
|
mode: ModelConnectionMode;
|
|
/** `null` before anything has been picked — the step opens this way. */
|
|
selectedId: string | null;
|
|
onSelect: (id: string) => void;
|
|
label: string;
|
|
/**
|
|
* Show only the chosen source, centred.
|
|
*
|
|
* The row is a question, and once a sign-in is running it has been answered —
|
|
* leaving the alternative on screen invites a press that would have to cancel
|
|
* a live server session to honour. Collapsing says the choice is made without
|
|
* disabling anything, which reads better than a greyed-out tile.
|
|
*/
|
|
collapsed?: boolean;
|
|
/**
|
|
* The row is returning to its default. Only changes how long the selected
|
|
* styling takes to leave — see the tile's own note.
|
|
*/
|
|
settling?: boolean;
|
|
}) {
|
|
const tiles = useRef(new Map<string, HTMLButtonElement>());
|
|
|
|
/**
|
|
* Arrow keys move the selection and the focus together, which is what a
|
|
* radio group is expected to do — without it the role would be announced and
|
|
* then not behave, which is worse than plain buttons. Selection wraps at both
|
|
* ends; three tiles is short enough that stopping at the edges just reads as
|
|
* the key having failed.
|
|
*/
|
|
const moveSelection = (delta: number) => {
|
|
if (sources.length === 0) return;
|
|
const current = sources.findIndex((source) => source.id === selectedId);
|
|
// Nothing picked yet: either arrow enters the row from the near end.
|
|
const from = current === -1 ? (delta > 0 ? -1 : 0) : current;
|
|
const next = (from + delta + sources.length) % sources.length;
|
|
const target = sources[next]!;
|
|
onSelect(target.id);
|
|
tiles.current.get(target.id)?.focus();
|
|
};
|
|
|
|
const shown = collapsed ? sources.filter((source) => source.id === selectedId) : sources;
|
|
|
|
return (
|
|
<div
|
|
role="radiogroup"
|
|
aria-label={label}
|
|
className={cn("flex items-start gap-3", collapsed && "justify-center")}
|
|
onKeyDown={(event) => {
|
|
// Collapsed, the row is a statement rather than a choice; arrow keys
|
|
// would move a selection that is no longer being asked for.
|
|
if (collapsed) return;
|
|
if (event.key === "ArrowRight" || event.key === "ArrowDown") {
|
|
event.preventDefault();
|
|
moveSelection(1);
|
|
} else if (event.key === "ArrowLeft" || event.key === "ArrowUp") {
|
|
event.preventDefault();
|
|
moveSelection(-1);
|
|
}
|
|
}}
|
|
>
|
|
{/*
|
|
`popLayout` takes the leaving tile out of flow at once, so the survivor's
|
|
`layout` animation targets its final centred position rather than
|
|
chasing a gap that is still closing.
|
|
|
|
The wrapper carries the width, not the tile: held at the width it had
|
|
with two in the row, so the kept tile travels without also growing.
|
|
*/}
|
|
<AnimatePresence initial={false} mode="popLayout">
|
|
{shown.map((source) => (
|
|
<motion.div
|
|
key={source.id}
|
|
layout
|
|
transition={SOURCE_COLLAPSE_MOVE}
|
|
exit={{ opacity: 0, transition: SOURCE_EXIT_FADE }}
|
|
className={cn(
|
|
"flex min-w-0",
|
|
collapsed ? "w-(--sz-source-tile-two-up)" : "flex-1",
|
|
)}
|
|
>
|
|
<ModelSourceTile
|
|
source={source}
|
|
mode={source.credentialMode === undefined ? mode : source.credentialMode}
|
|
selected={source.id === selectedId}
|
|
onSelect={() => onSelect(source.id)}
|
|
settling={settling}
|
|
buttonRef={(node) => {
|
|
if (node) tiles.current.set(source.id, node);
|
|
else tiles.current.delete(source.id);
|
|
}}
|
|
/>
|
|
</motion.div>
|
|
))}
|
|
</AnimatePresence>
|
|
</div>
|
|
);
|
|
}
|