Files
paperclip/ui/src/components/TrustPresetSection.tsx
T
Dotta dbebf30c89 Add low-trust review containment (#7530)
## Thinking Path

> - Paperclip is a control plane for AI-agent companies, so execution
policy and trust boundaries are part of the product's safety contract.
> - Low-trust review work needs narrower authority than normal
same-company agents because hostile PRs, comments, attachments, and
generated output can carry prompt-injection payloads.
> - The current V1 shape gives trusted workers broad company context,
which is useful for normal execution but too permissive for a reviewer
assigned to hostile content.
> - This branch adds a `low_trust_review` preset, source-trust tagging,
route-level containment, and quarantine handling so low-trust output
does not automatically flow into higher-trust wake context.
> - The branch has been rebased onto current `origin/master`, and the
low-trust migration was renumbered to `0097_low_trust_source_trust.sql`
to avoid collisions with existing `0091` through `0096` migrations.
> - Greptile feedback was addressed by tightening low-trust detection,
preserving project-level trust policy checks, fixing issue-kind
promotion lookup, removing duplicate post-lease isolation assertion,
documenting fail-closed source-trust behavior, bounding ancestry checks,
enforcing runtime issue context for CEOs, awaiting accepted-plan monitor
authorization, and making low-trust issue source-trust tagging atomic.
> - The benefit is a first production slice of deny-by-default review
containment with regression coverage for the main control-plane pivot
surfaces.

Fixes #7531.

## What Changed

- Added shared trust-policy types and validators, plus
database/source-trust fields for issues, comments, documents, and work
products.
- Implemented server enforcement for low-trust issue scope, agent
self-view redaction, secret/plugin/runtime denial paths, promotion
checks, and quarantined continuation/wake context.
- Added focused low-trust regression tests for resolver behavior, source
trust, route authorization, heartbeat preflight ordering, runtime
containment, and quarantine redaction.
- Added board UI affordances for selecting/reviewing the low-trust
preset and surfacing source-trust badges in relevant issue views.
- Added `doc/LOW-TRUST-PRESETS.md`, updated
`doc/SPEC-implementation.md`, and committed the low-trust review
contract plan under `doc/plans/`.
- Rebasing note: the original `0097_low_trust_source_trust.sql`
migration was renamed to `0097_low_trust_source_trust.sql`; the SQL uses
`ADD COLUMN IF NOT EXISTS` so users who already applied the old-numbered
migration are not broken by the renumbered migration.

## Verification

- Rebased branch onto current `origin/master` and force-pushed with
lease to `origin/PAP-10211-low-trust-agent` at head `2719f31e3`.
- Confirmed the PR diff does not include `pnpm-lock.yaml` or
`.github/workflows` changes.
- Resolved upstream UI/comment conflicts by preserving deleted-comment
tombstone behavior and low-trust source-trust badges/metadata.
- Renumbered the low-trust source-trust migration to
`0097_low_trust_source_trust.sql`; the SQL uses `ADD COLUMN IF NOT
EXISTS` so users who already applied an old-numbered copy are not
broken.
- `pnpm exec vitest run ui/src/lib/issue-chat-messages.test.ts
server/src/__tests__/heartbeat-workspace-session.test.ts`
- `pnpm exec vitest run server/src/__tests__/source-trust.test.ts
server/src/__tests__/workspace-runtime-service-authz.test.ts
ui/src/lib/trust-policy-ui.test.ts
ui/src/components/TrustPresetSection.test.tsx`
- `pnpm run typecheck:build-gaps`
- `git diff --check`
- GitHub checks pass on head `2719f31e3`: build, typecheck/release
registry, general tests, serialized server suites, e2e, canary, verify,
policy/review, Socket, and Snyk.
- Greptile Review passes with Confidence Score 5/5 and zero unresolved
Greptile review threads.
- No design screenshots/images were added because the task explicitly
says not to add them unless they are specifically part of the work.

## Risks

- Medium risk: this touches shared trust-policy contracts, server
authorization paths, heartbeat context generation, migration metadata,
and UI preset controls.
- Low-trust containment is intentionally deny-by-default; legitimate
future review workflows may need explicit allowlisted exceptions.
- Plugin/runtime/security surfaces are broad, so regression tests cover
the current known routes but future integrations must route through the
same containment layer.
- The PR is ready for review; GitHub checks are green and Greptile is
5/5.

> 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

- OpenAI Codex, GPT-5 coding agent, tool-enabled shell and GitHub CLI
workflow.

## 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 run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] UI changes are covered by focused tests; no screenshots were added
per task instruction not to add design images unless specifically
required
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge

---------

Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-06-05 16:48:02 -05:00

259 lines
11 KiB
TypeScript

import { useEffect, useMemo, useState } from "react";
import type { AgentPermissions, TrustPreset } from "@paperclipai/shared";
import { Lock, ShieldAlert } from "lucide-react";
import { Button } from "@/components/ui/button";
import { Field, CollapsibleSection } from "./agent-config-primitives";
import {
buildPermissionsForTrustPreset,
clearSingleLowTrustBoundaryTarget,
getLowTrustBoundary,
getSingleLowTrustBoundaryTarget,
getTrustPreset,
isCeLowTrustBoundaryEditable,
lowTrustBoundaryHasScope,
setSingleLowTrustBoundaryTarget,
summarizeLowTrustBoundaryTarget,
TRUST_PRESET_DESCRIPTIONS,
TRUST_PRESET_LABELS,
type LowTrustBoundaryTarget,
} from "../lib/trust-policy-ui";
import { cn } from "../lib/utils";
const inputClass =
"w-full rounded-md border border-border px-2.5 py-1.5 bg-transparent outline-none text-sm font-mono placeholder:text-muted-foreground/40";
function formatCount(value: readonly unknown[] | undefined, singular: string, plural: string) {
const count = value?.length ?? 0;
if (count === 0) return "-";
return `${count} ${count === 1 ? singular : plural}`;
}
function PolicyRow({ label, value }: { label: string; value: string }) {
return (
<div className="flex items-start justify-between gap-3 py-1.5 text-sm">
<span className="shrink-0 text-xs text-muted-foreground">{label}</span>
<span className={cn("min-w-0 text-right", value === "-" && "text-muted-foreground")}>{value}</span>
</div>
);
}
export interface LowTrustBoundaryCandidate {
id: string;
label: string;
}
type LowTrustBoundaryTargetType = LowTrustBoundaryTarget["type"];
const BOUNDARY_TARGET_LABELS: Record<LowTrustBoundaryTargetType, string> = {
project: "Project",
root_issue: "Root issue",
issue: "Issue",
};
export function TrustPresetSection({
permissions,
onChange,
disabled,
companyId,
projectCandidates = [],
issueCandidates = [],
candidatesLoading,
}: {
permissions: Partial<AgentPermissions> | null | undefined;
onChange: (permissions: Partial<AgentPermissions>) => void;
disabled?: boolean;
companyId?: string | null;
projectCandidates?: LowTrustBoundaryCandidate[];
issueCandidates?: LowTrustBoundaryCandidate[];
candidatesLoading?: boolean;
}) {
const [policyOpen, setPolicyOpen] = useState(false);
const preset = getTrustPreset(permissions);
const boundary = getLowTrustBoundary(permissions);
const boundaryTarget = getSingleLowTrustBoundaryTarget(boundary);
const [targetType, setTargetType] = useState<LowTrustBoundaryTargetType>(boundaryTarget?.type ?? "project");
const lowTrust = preset === "low_trust_review";
const hasScope = lowTrustBoundaryHasScope(boundary);
const boundaryEditable = isCeLowTrustBoundaryEditable(boundary);
const policy = permissions?.authorizationPolicy ?? null;
const managedPermissions = useMemo(
() => buildPermissionsForTrustPreset(permissions, preset),
[permissions, preset],
);
useEffect(() => {
if (boundaryTarget) setTargetType(boundaryTarget.type);
}, [boundaryTarget?.type]);
function handlePresetChange(value: string) {
const nextPreset: TrustPreset = value === "low_trust_review" ? "low_trust_review" : "standard";
onChange(buildPermissionsForTrustPreset(permissions, nextPreset));
}
function handleBoundaryTargetChange(targetId: string) {
if (!companyId || !targetId) return;
onChange(setSingleLowTrustBoundaryTarget(permissions, companyId, { type: targetType, id: targetId }));
}
function handleClearBoundary() {
onChange(clearSingleLowTrustBoundaryTarget(permissions));
}
const targetCandidates = targetType === "project" ? projectCandidates : issueCandidates;
const boundaryValue = boundaryTarget?.type === targetType ? boundaryTarget.id : "";
return (
<div>
<h3 className="mb-3 text-sm font-medium">Trust</h3>
<div className="rounded-lg border border-border p-4 space-y-3">
<Field label="Trust preset" hint="Choose how broadly this agent can read and act on Paperclip work objects.">
<select
className={inputClass}
value={preset}
onChange={(event) => handlePresetChange(event.target.value)}
disabled={disabled}
>
<option value="standard">{TRUST_PRESET_LABELS.standard}</option>
<option value="low_trust_review">{TRUST_PRESET_LABELS.low_trust_review}</option>
</select>
</Field>
<p className="text-xs text-muted-foreground">{TRUST_PRESET_DESCRIPTIONS[preset]}</p>
{lowTrust ? (
<div
role={hasScope ? "status" : "alert"}
aria-live="polite"
className={cn(
"rounded-md border px-3 py-2.5 text-sm flex gap-2",
hasScope
? "border-amber-500/30 bg-amber-500/10 text-amber-700 dark:text-amber-100"
: "border-destructive/30 bg-destructive/10 text-destructive",
)}
>
{hasScope ? (
<ShieldAlert className="mt-0.5 h-4 w-4 shrink-0" />
) : (
<Lock className="mt-0.5 h-4 w-4 shrink-0" />
)}
<div className="min-w-0 flex-1 space-y-2">
<div>
<p className="font-medium">
{hasScope ? "Containment active" : "Containment not configured"}
</p>
<p className="mt-1 text-xs leading-5">
{hasScope
? "This agent can only read and mutate work inside its assigned review boundary. Raw output is quarantined from higher-trust agents until a trusted reviewer promotes it."
: "This agent is set to low-trust review, but no project, root issue, or issue scope is set in the core policy. Add a scope before this agent can run without denial."}
</p>
</div>
{boundaryEditable ? (
<div className="rounded-md border border-border/70 bg-background/70 p-3 text-foreground space-y-3">
<div className="grid gap-3 sm:grid-cols-[minmax(0,0.75fr)_minmax(0,1fr)]">
<Field label="Boundary type">
<select
className={inputClass}
value={targetType}
onChange={(event) => setTargetType(event.target.value as LowTrustBoundaryTargetType)}
disabled={disabled}
>
<option value="project">Project</option>
<option value="root_issue">Root issue</option>
<option value="issue">Issue</option>
</select>
</Field>
<Field label={BOUNDARY_TARGET_LABELS[targetType]}>
<select
className={inputClass}
value={boundaryValue}
onChange={(event) => handleBoundaryTargetChange(event.target.value)}
disabled={disabled || !companyId || candidatesLoading || targetCandidates.length === 0}
>
<option value="">
{candidatesLoading
? "Loading…"
: targetCandidates.length === 0
? `No ${targetType === "project" ? "projects" : "issues"} available`
: "Select boundary"}
</option>
{targetCandidates.map((candidate) => (
<option key={candidate.id} value={candidate.id}>
{candidate.label}
</option>
))}
</select>
</Field>
</div>
<div className="flex flex-wrap items-center justify-between gap-2">
<p className="text-xs text-muted-foreground">
CE saves one containment boundary at a time. Saved policies include this company id.
</p>
{boundaryTarget ? (
<Button
type="button"
size="sm"
variant="outline"
className="h-7 px-2.5 text-xs"
onClick={handleClearBoundary}
disabled={disabled}
>
Clear boundary
</Button>
) : null}
</div>
</div>
) : (
<div className="rounded-md border border-border/70 bg-background/70 p-3 text-foreground">
<p className="text-sm font-medium">Managed by EE/API</p>
<p className="mt-1 text-xs leading-5 text-muted-foreground">
This policy has {summarizeLowTrustBoundaryTarget(boundary).toLowerCase()} and cannot be edited by the CE single-boundary editor.
</p>
</div>
)}
<p className="text-xs text-muted-foreground">
Want to set more than one containment boundary?{" "}
<a
className="underline underline-offset-2 hover:text-foreground"
href="https://paperclip.ing/ee"
target="_blank"
rel="noreferrer"
>
Get Paperclip EE.
</a>
</p>
<CollapsibleSection
title="View policy"
open={policyOpen}
onToggle={() => setPolicyOpen((open) => !open)}
>
<div className="divide-y divide-border/60 text-foreground">
<PolicyRow label="Preset" value="Low-trust review v1" />
<PolicyRow label="Raw output" value="Quarantined from higher-trust agents" />
<PolicyRow label="Projects" value={formatCount(boundary?.projectIds, "project", "projects")} />
<PolicyRow label="Root issue" value={boundary?.rootIssueId ? boundary.rootIssueId.slice(0, 8) : "-"} />
<PolicyRow label="Explicit issues" value={formatCount(boundary?.issueIds, "issue", "issues")} />
<PolicyRow label="Allowed agents" value={formatCount(boundary?.allowedAgentIds, "agent", "agents")} />
<PolicyRow label="Allowed tools" value={boundary?.allowedToolClasses?.join(" · ") || "-"} />
<PolicyRow label="Allowed secrets" value={formatCount(boundary?.allowedSecretBindingIds, "binding", "bindings")} />
<PolicyRow label="Promotion target" value={boundary?.outputPromotionTarget?.issueId?.slice(0, 8) ?? "-"} />
<PolicyRow
label="EE fields"
value={Object.keys(policy ?? {}).some((key) => !["trustPreset", "reviewPreset", "trustBoundary"].includes(key))
? "Custom advanced policy fields preserved"
: "-"}
/>
</div>
</CollapsibleSection>
</div>
</div>
) : null}
{managedPermissions.authorizationPolicy?.reviewPreset ? null : (
<p className="text-xs text-muted-foreground">
Advanced permissions remain editable through the EE permissions extension when installed.
</p>
)}
</div>
</div>
);
}