mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-10-21 00:05:49 +02:00
## Thinking Path > - Paperclip runs AI agents through qualified provider backends. > - The direct live eval workflow builds one immutable Runner runtime for every matrix cell. > - The workflow reinstalled the packed Runner with npm. > - That install discarded pnpm patches and selected provider dependencies outside the qualified lock. > - The first pnpm deployment model also placed its virtual-store marker at the wrong level; a real deployment keeps `.pnpm` beside the scoped Runner package. > - AgentCore enforced the current context-aware harness but the direct eval CLI did not supply the production v3 runtime context that harness requires. > - This pull request preserves the qualified dependency graph, resolves the real deployment layout, and makes direct evals exercise the production runtime-context contract. > - The benefit is that live eval cells reach their provider turn with the same artifacts and context contract that Paperclip qualified. ## Linked Issues or Issue Description Refs: #12931 **What happened?** The full direct live eval campaign failed every ACPX cell during `session.open`. The portable runtime had an incorrect dependency root. Its npm install also discarded the qualified ACP server patches. AgentCore cells first failed because Runner enforced `aws-agentcore-harness-v1` while the provisioned stack and eval profile use `aws-agentcore-harness-context-v2`; after aligning that revision, the direct eval CLI still omitted the required v3 runtime context. **Expected behavior** The direct eval runtime must preserve the frozen pnpm dependency graph and patched provider bytes. Runner, server validation, OpenAPI, and the deployed AgentCore stack must use one qualification revision. Direct eval attempts must supply the same immutable native runtime-context contract as production. **Steps to reproduce** 1. Dispatch `Runner Direct Live Protocol Evals` from `master`. 2. Select an ACPX Claude, ACPX Codex, or AgentCore roster. 3. Observe a pre-turn provider bootstrap failure. **Paperclip version or commit** `d96452db059338b329b458ba8fe359fef72f1363` **Deployment mode** GitHub Actions on the RunsOn Linux x64 fleet. ## What Changed - Build the reusable direct-eval runtime with `pnpm deploy --prod`. - Resolve ACPX dependencies from the actual scoped-package layout of a self-contained pnpm deployment. - Align AgentCore configuration and qualification checks on `aws-agentcore-harness-context-v2`. - Materialize a minimal immutable v3 runtime context for each isolated direct eval attempt. - Add workflow, package-authority, runtime-context, Rust, and server regression coverage. - Document the qualified packaging, runtime-context, and AgentCore revision contracts. ## Verification - `pnpm --filter @paperclipai/paperclip-runner typecheck` - `pnpm --filter @paperclipai/server typecheck` - `pnpm --filter @paperclipai/paperclip-runner exec vitest run src/live/runnerd-codex-transport.test.ts` (70 tests) - `pnpm --filter @paperclipai/paperclip-runner exec vitest run src/cli/eval-session-contract.test.ts` (14 tests) - Focused Runner contract tests (36 tests) - Focused server profile tests (47 tests) - Focused Rust managed-provider and native-selector tests (19 tests) - `node --test packages/paperclip-runner/scripts/runner-protocol-eval-workflow-security.test.mjs` - `actionlint .github/workflows/runner-protocol-live-evals.yml` - A local `pnpm deploy --prod` produced both qualified ACP server digests. - A Linux reproduction of the first follow-up smoke identified the real deployment root and the missing AgentCore runtime context. ## Risks The AgentCore revision change rejects profiles that still use the obsolete v1 value. This is intentional because the provisioned context-aware harness and current eval profile use v2. Direct eval prompts now receive the same fixed runtime-context preamble as production, so behavior scores may move; that is the intended qualification surface. The workflow package layout changes, but tests assert the new entrypoint and dependency root. This change does not modify the browser full-stack E2E workflow. > 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. See `CONTRIBUTING.md`. ## Model Used OpenAI Codex with GPT-5.6. The context-window size is not exposed in this session. The model used extended reasoning, repository tools, code execution, Docker-based Linux reproduction, and GitHub Actions diagnostics. ## 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 (for example, `docs/...` or `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 --------- Co-authored-by: Paperclip <noreply@paperclip.ing>
576 lines
19 KiB
TypeScript
576 lines
19 KiB
TypeScript
import type { Db } from "@paperclipai/db";
|
|
import { describe, expect, it } from "vitest";
|
|
|
|
import {
|
|
computeManagedAgentProfileRevision,
|
|
managedAgentProfileService,
|
|
} from "./managed-agent-profiles.js";
|
|
import {
|
|
computeRemoteAgentProfileRevision,
|
|
remoteAgentProfileService,
|
|
type RemoteAgentProfileInput,
|
|
} from "./remote-agent-profiles.js";
|
|
|
|
const COMPANY_ID = "10000000-0000-4000-8000-000000000001";
|
|
const OTHER_COMPANY_SECRET_ID = "20000000-0000-4000-8000-000000000002";
|
|
|
|
const AWS_CONFIGURATION = {
|
|
region: "us-east-1",
|
|
accountId: "123456789012",
|
|
harnessArn: "arn:aws:bedrock-agentcore:us-east-1:123456789012:harness/example",
|
|
harnessVersion: "1",
|
|
endpointArn: "arn:aws:bedrock-agentcore:us-east-1:123456789012:endpoint/example",
|
|
endpointQualifier: "paperclip",
|
|
agentRuntimeArn: "arn:aws:bedrock-agentcore:us-east-1:123456789012:runtime/example",
|
|
memoryArn: "arn:aws:bedrock-agentcore:us-east-1:123456789012:memory/example",
|
|
memoryId: "memory-example",
|
|
invocationRoleArn: "arn:aws:iam::123456789012:role/paperclip-runner",
|
|
contextBucket: "paperclip-runner-context",
|
|
contextPrefix: "profiles/example",
|
|
contextKmsKeyArn: "arn:aws:kms:us-east-1:123456789012:key/example",
|
|
qualificationRevision: "aws-agentcore-harness-context-v2",
|
|
defaultModel: "global.anthropic.claude-sonnet-4-6",
|
|
eventExpiryDays: 90,
|
|
defaultMaxEstimatedSessionCostUsd: 1,
|
|
} as const;
|
|
|
|
function remoteInput(
|
|
overrides: Partial<RemoteAgentProfileInput> = {},
|
|
): RemoteAgentProfileInput {
|
|
return {
|
|
profileKey: "agentcore",
|
|
displayName: "AgentCore",
|
|
service: "aws_bedrock_agentcore_harness",
|
|
configuration: { ...AWS_CONFIGURATION },
|
|
enabled: false,
|
|
retentionAcknowledged: false,
|
|
qualification: { suite: "aws-agentcore-harness-context-v2" },
|
|
...overrides,
|
|
};
|
|
}
|
|
|
|
function dbReturningNoRows(): Db {
|
|
return {
|
|
select: () => ({
|
|
from: () => ({
|
|
where: () => ({
|
|
limit: async () => [],
|
|
}),
|
|
}),
|
|
}),
|
|
} as unknown as Db;
|
|
}
|
|
|
|
function dbReturningFirstRow(row: Record<string, unknown>): Db {
|
|
return {
|
|
select: () => ({
|
|
from: () => ({
|
|
where: () => ({
|
|
limit: async () => [row],
|
|
}),
|
|
}),
|
|
}),
|
|
} as unknown as Db;
|
|
}
|
|
|
|
function dbForUpsert(
|
|
existing: Record<string, unknown> | null,
|
|
secretId?: string,
|
|
): Db {
|
|
let selection = 0;
|
|
return {
|
|
select: () => ({
|
|
from: () => ({
|
|
where: () => ({
|
|
limit: async () => {
|
|
selection += 1;
|
|
if (selection === 1) return existing ? [existing] : [];
|
|
return secretId ? [{ id: secretId }] : [];
|
|
},
|
|
}),
|
|
}),
|
|
}),
|
|
insert: () => ({
|
|
values: (values: Record<string, unknown>) => ({
|
|
onConflictDoUpdate: () => ({
|
|
returning: async () => [{ ...(existing ?? {}), ...values }],
|
|
}),
|
|
}),
|
|
}),
|
|
} as unknown as Db;
|
|
}
|
|
|
|
const unusedDb = new Proxy({}, {
|
|
get() {
|
|
throw new Error("validation unexpectedly accessed the database");
|
|
},
|
|
}) as unknown as Db;
|
|
|
|
describe("remote agent profile metadata validation", () => {
|
|
it("rejects obsolete explicit credentials because AgentCore uses workload identity", async () => {
|
|
await expect(
|
|
remoteAgentProfileService(unusedDb).upsert(COMPANY_ID, {
|
|
...remoteInput(),
|
|
credentialSecretId: OTHER_COMPANY_SECRET_ID,
|
|
} as unknown as RemoteAgentProfileInput),
|
|
).rejects.toThrow("use workload identity");
|
|
});
|
|
|
|
it("rejects Claude profiles before the remote AgentCore store is accessed", async () => {
|
|
await expect(
|
|
remoteAgentProfileService(unusedDb).upsert(COMPANY_ID, {
|
|
...remoteInput(),
|
|
service: "anthropic_managed_agents",
|
|
} as unknown as RemoteAgentProfileInput),
|
|
).rejects.toThrow("Unsupported remote agent service");
|
|
});
|
|
|
|
it("rejects non-canonical Anthropic Agent versions before profile storage", async () => {
|
|
for (const agentVersion of ["latest", "0", "01", "2147483648"]) {
|
|
await expect(
|
|
managedAgentProfileService(unusedDb).upsert(COMPANY_ID, {
|
|
profileKey: "managed",
|
|
displayName: "Managed Agent",
|
|
anthropicAgentId: "agent-example",
|
|
agentVersion,
|
|
environmentId: "environment-example",
|
|
defaultModel: "claude-sonnet-5",
|
|
defaultMaxListCostUsd: 1,
|
|
apiKeySecretId: OTHER_COMPANY_SECRET_ID,
|
|
enabled: false,
|
|
retentionAcknowledged: false,
|
|
}),
|
|
).rejects.toThrow("canonical positive 32-bit integer");
|
|
}
|
|
});
|
|
|
|
it("rejects provider configuration keys outside the exact allowlist", async () => {
|
|
await expect(
|
|
remoteAgentProfileService(unusedDb).upsert(
|
|
COMPANY_ID,
|
|
remoteInput({
|
|
configuration: {
|
|
...AWS_CONFIGURATION,
|
|
customEndpointToken: "not-persistable",
|
|
},
|
|
}),
|
|
),
|
|
).rejects.toThrow("Unsupported configuration field");
|
|
});
|
|
|
|
it("rejects secret-shaped configuration values and qualification fields", async () => {
|
|
await expect(
|
|
remoteAgentProfileService(unusedDb).upsert(
|
|
COMPANY_ID,
|
|
remoteInput({
|
|
configuration: {
|
|
...AWS_CONFIGURATION,
|
|
qualificationRevision: "Bearer secret-value",
|
|
},
|
|
}),
|
|
),
|
|
).rejects.toThrow("must not contain credential-shaped keys or values");
|
|
|
|
await expect(
|
|
remoteAgentProfileService(unusedDb).upsert(
|
|
COMPANY_ID,
|
|
remoteInput({ qualification: { apiToken: "secret-value" } }),
|
|
),
|
|
).rejects.toThrow("must not contain credential-shaped keys or values");
|
|
|
|
await expect(
|
|
managedAgentProfileService(unusedDb).upsert(COMPANY_ID, {
|
|
profileKey: "managed",
|
|
displayName: "Bearer secret-value",
|
|
anthropicAgentId: "agent-example",
|
|
agentVersion: "1",
|
|
environmentId: "environment-example",
|
|
defaultModel: "claude-sonnet-5",
|
|
defaultMaxListCostUsd: 1,
|
|
apiKeySecretId: OTHER_COMPANY_SECRET_ID,
|
|
enabled: false,
|
|
retentionAcknowledged: false,
|
|
}),
|
|
).rejects.toThrow("must not contain credential-shaped keys or values");
|
|
});
|
|
|
|
it("requires retention acknowledgement and a positive explicit spend ceiling", async () => {
|
|
await expect(
|
|
remoteAgentProfileService(unusedDb).upsert(
|
|
COMPANY_ID,
|
|
remoteInput({ enabled: true, retentionAcknowledged: false }),
|
|
),
|
|
).rejects.toThrow("requires retention acknowledgement");
|
|
|
|
await expect(
|
|
remoteAgentProfileService(unusedDb).upsert(
|
|
COMPANY_ID,
|
|
remoteInput({
|
|
configuration: {
|
|
...AWS_CONFIGURATION,
|
|
defaultMaxEstimatedSessionCostUsd: 0,
|
|
},
|
|
}),
|
|
),
|
|
).rejects.toThrow("spend ceiling must be positive");
|
|
|
|
const {
|
|
defaultMaxEstimatedSessionCostUsd: _omittedSpendCeiling,
|
|
...configurationWithoutSpendCeiling
|
|
} = AWS_CONFIGURATION;
|
|
await expect(
|
|
remoteAgentProfileService(unusedDb).upsert(
|
|
COMPANY_ID,
|
|
remoteInput({ configuration: configurationWithoutSpendCeiling }),
|
|
),
|
|
).rejects.toThrow("spend ceiling must be positive");
|
|
|
|
await expect(
|
|
managedAgentProfileService(unusedDb).upsert(COMPANY_ID, {
|
|
profileKey: "managed",
|
|
displayName: "Managed Agent",
|
|
anthropicAgentId: "agent-example",
|
|
agentVersion: "1",
|
|
environmentId: "environment-example",
|
|
defaultModel: "claude-sonnet-5",
|
|
defaultMaxListCostUsd: 0,
|
|
apiKeySecretId: OTHER_COMPANY_SECRET_ID,
|
|
enabled: false,
|
|
retentionAcknowledged: false,
|
|
}),
|
|
).rejects.toThrow("spend ceiling must be positive");
|
|
});
|
|
|
|
it("does not treat enablement or arbitrary metadata as qualification", async () => {
|
|
await expect(
|
|
remoteAgentProfileService(unusedDb).upsert(
|
|
COMPANY_ID,
|
|
remoteInput({ enabled: true, retentionAcknowledged: true, qualification: {} }),
|
|
),
|
|
).rejects.toThrow("qualification attestation is required");
|
|
|
|
await expect(
|
|
remoteAgentProfileService(unusedDb).upsert(
|
|
COMPANY_ID,
|
|
remoteInput({
|
|
enabled: true,
|
|
retentionAcknowledged: true,
|
|
qualification: { suite: "operator-says-pass" },
|
|
}),
|
|
),
|
|
).rejects.toThrow("does not match the qualified harness suite");
|
|
|
|
await expect(
|
|
managedAgentProfileService(unusedDb).upsert(COMPANY_ID, {
|
|
profileKey: "managed",
|
|
displayName: "Managed Agent",
|
|
anthropicAgentId: "agent-example",
|
|
agentVersion: "1",
|
|
environmentId: "environment-example",
|
|
defaultModel: "claude-sonnet-5",
|
|
defaultMaxListCostUsd: 1,
|
|
apiKeySecretId: OTHER_COMPANY_SECRET_ID,
|
|
enabled: true,
|
|
retentionAcknowledged: true,
|
|
qualification: {},
|
|
}),
|
|
).rejects.toThrow("qualification attestation is required");
|
|
});
|
|
|
|
it("stores a qualified revision only after exact operator attestation", async () => {
|
|
const qualification = {
|
|
probedAt: "2026-08-01T00:00:00.000Z",
|
|
betaVersion: "managed-agents-2026-04-01",
|
|
environmentPolicy: "limited_no_hosts_no_packages",
|
|
agentCapabilities: "no_tools_no_mcp_no_skills_no_multiagent",
|
|
};
|
|
const managed = await managedAgentProfileService(
|
|
dbForUpsert(null, OTHER_COMPANY_SECRET_ID),
|
|
).upsert(COMPANY_ID, {
|
|
profileKey: "managed",
|
|
displayName: "Managed Agent",
|
|
anthropicAgentId: "agent-example",
|
|
agentVersion: "1",
|
|
environmentId: "environment-example",
|
|
defaultModel: "claude-sonnet-5",
|
|
defaultMaxListCostUsd: 1,
|
|
apiKeySecretId: OTHER_COMPANY_SECRET_ID,
|
|
enabled: true,
|
|
retentionAcknowledged: true,
|
|
qualification,
|
|
});
|
|
expect(managed.qualifiedAt).toBeInstanceOf(Date);
|
|
expect(managed.qualifiedRevision).toMatch(/^sha256:[0-9a-f]{64}$/);
|
|
|
|
const remote = await remoteAgentProfileService(dbForUpsert(null)).upsert(
|
|
COMPANY_ID,
|
|
remoteInput({ enabled: true, retentionAcknowledged: true }),
|
|
);
|
|
expect(remote.qualifiedAt).toBeInstanceOf(Date);
|
|
expect(remote.qualifiedRevision).toMatch(/^sha256:[0-9a-f]{64}$/);
|
|
});
|
|
|
|
it("requires exact qualified models and non-UUID profile keys", async () => {
|
|
await expect(
|
|
remoteAgentProfileService(unusedDb).upsert(
|
|
COMPANY_ID,
|
|
remoteInput({
|
|
configuration: { ...AWS_CONFIGURATION, defaultModel: "claude-sonnet-4-6" },
|
|
}),
|
|
),
|
|
).rejects.toThrow("model must be global.anthropic.claude-sonnet-4-6");
|
|
|
|
await expect(
|
|
managedAgentProfileService(unusedDb).upsert(COMPANY_ID, {
|
|
profileKey: "managed",
|
|
displayName: "Managed Agent",
|
|
anthropicAgentId: "agent-example",
|
|
agentVersion: "1",
|
|
environmentId: "environment-example",
|
|
defaultModel: "claude-opus-5",
|
|
defaultMaxListCostUsd: 1,
|
|
apiKeySecretId: OTHER_COMPANY_SECRET_ID,
|
|
enabled: false,
|
|
retentionAcknowledged: false,
|
|
}),
|
|
).rejects.toThrow("model must be claude-sonnet-5");
|
|
|
|
await expect(
|
|
remoteAgentProfileService(unusedDb).upsert(
|
|
COMPANY_ID,
|
|
remoteInput({ profileKey: "30000000-0000-4000-8000-000000000003" }),
|
|
),
|
|
).rejects.toThrow("must not be UUID-shaped");
|
|
});
|
|
|
|
it("rejects Claude credential references that do not resolve in the owning company", async () => {
|
|
await expect(
|
|
managedAgentProfileService(dbReturningNoRows()).upsert(COMPANY_ID, {
|
|
profileKey: "managed",
|
|
displayName: "Managed Agent",
|
|
anthropicAgentId: "agent-example",
|
|
agentVersion: "1",
|
|
environmentId: "environment-example",
|
|
defaultModel: "claude-sonnet-5",
|
|
defaultMaxListCostUsd: 1,
|
|
apiKeySecretId: OTHER_COMPANY_SECRET_ID,
|
|
enabled: false,
|
|
retentionAcknowledged: false,
|
|
qualification: {},
|
|
}),
|
|
).rejects.toThrow("API-key secret reference is invalid");
|
|
});
|
|
|
|
it("does not allow a qualified AgentCore profile key to be repointed", async () => {
|
|
const qualification = { suite: "aws-agentcore-harness-context-v2" };
|
|
const existing = {
|
|
id: "30000000-0000-4000-8000-000000000003",
|
|
companyId: COMPANY_ID,
|
|
profileKey: "agentcore",
|
|
displayName: "AgentCore",
|
|
service: "aws_bedrock_agentcore_harness",
|
|
configuration: { ...AWS_CONFIGURATION },
|
|
enabled: true,
|
|
retentionAcknowledged: true,
|
|
qualification,
|
|
qualifiedAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
qualifiedRevision: computeRemoteAgentProfileRevision({
|
|
service: "aws_bedrock_agentcore_harness",
|
|
configuration: AWS_CONFIGURATION,
|
|
retentionAcknowledged: true,
|
|
qualification,
|
|
}),
|
|
createdAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
updatedAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
};
|
|
|
|
await expect(
|
|
remoteAgentProfileService(dbReturningFirstRow(existing)).upsert(
|
|
COMPANY_ID,
|
|
remoteInput({
|
|
enabled: true,
|
|
retentionAcknowledged: true,
|
|
configuration: { ...AWS_CONFIGURATION, memoryId: "different-memory" },
|
|
}),
|
|
),
|
|
).rejects.toThrow("configuration revision is immutable");
|
|
});
|
|
|
|
it("does not allow qualified Claude resource identity or proof to drift", async () => {
|
|
const qualification = {
|
|
probedAt: "2026-08-01T00:00:00.000Z",
|
|
betaVersion: "managed-agents-2026-04-01",
|
|
environmentPolicy: "limited_no_hosts_no_packages",
|
|
agentCapabilities: "no_tools_no_mcp_no_skills_no_multiagent",
|
|
};
|
|
const existing = {
|
|
id: "30000000-0000-4000-8000-000000000004",
|
|
companyId: COMPANY_ID,
|
|
profileKey: "managed",
|
|
displayName: "Managed Agent",
|
|
service: "anthropic_managed_agents",
|
|
anthropicAgentId: "agent-example",
|
|
agentVersion: "1",
|
|
environmentId: "environment-example",
|
|
betaVersion: "managed-agents-2026-04-01",
|
|
defaultModel: "claude-sonnet-5",
|
|
defaultMaxListCostCents: 100,
|
|
apiKeySecretId: OTHER_COMPANY_SECRET_ID,
|
|
enabled: true,
|
|
retentionAcknowledged: true,
|
|
qualification,
|
|
qualifiedAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
qualifiedRevision: computeManagedAgentProfileRevision({
|
|
anthropicAgentId: "agent-example",
|
|
agentVersion: "1",
|
|
environmentId: "environment-example",
|
|
betaVersion: "managed-agents-2026-04-01",
|
|
retentionAcknowledged: true,
|
|
qualification,
|
|
}),
|
|
createdAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
updatedAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
};
|
|
|
|
await expect(
|
|
managedAgentProfileService(dbReturningFirstRow(existing)).upsert(COMPANY_ID, {
|
|
profileKey: "managed",
|
|
displayName: "Managed Agent",
|
|
anthropicAgentId: "different-agent",
|
|
agentVersion: "1",
|
|
environmentId: "environment-example",
|
|
defaultModel: "claude-sonnet-5",
|
|
defaultMaxListCostUsd: 1,
|
|
apiKeySecretId: OTHER_COMPANY_SECRET_ID,
|
|
enabled: true,
|
|
retentionAcknowledged: true,
|
|
qualification,
|
|
}),
|
|
).rejects.toThrow("configuration revision is immutable");
|
|
});
|
|
|
|
it("allows credential rotation and mutable caps without invalidating qualified identity", async () => {
|
|
const qualification = {
|
|
probedAt: "2026-08-01T00:00:00.000Z",
|
|
betaVersion: "managed-agents-2026-04-01",
|
|
environmentPolicy: "limited_no_hosts_no_packages",
|
|
agentCapabilities: "no_tools_no_mcp_no_skills_no_multiagent",
|
|
};
|
|
const rotatedSecretId = "20000000-0000-4000-8000-000000000003";
|
|
const existingManaged = {
|
|
id: "30000000-0000-4000-8000-000000000004",
|
|
companyId: COMPANY_ID,
|
|
profileKey: "managed",
|
|
displayName: "Managed Agent",
|
|
service: "anthropic_managed_agents",
|
|
anthropicAgentId: "agent-example",
|
|
agentVersion: "1",
|
|
environmentId: "environment-example",
|
|
betaVersion: "managed-agents-2026-04-01",
|
|
defaultModel: "claude-sonnet-5",
|
|
defaultMaxListCostCents: 100,
|
|
apiKeySecretId: OTHER_COMPANY_SECRET_ID,
|
|
enabled: true,
|
|
retentionAcknowledged: true,
|
|
qualification,
|
|
qualifiedAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
qualifiedRevision: computeManagedAgentProfileRevision({
|
|
anthropicAgentId: "agent-example",
|
|
agentVersion: "1",
|
|
environmentId: "environment-example",
|
|
betaVersion: "managed-agents-2026-04-01",
|
|
retentionAcknowledged: true,
|
|
qualification,
|
|
}),
|
|
createdAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
updatedAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
};
|
|
const updatedManaged = await managedAgentProfileService(
|
|
dbForUpsert(existingManaged, rotatedSecretId),
|
|
).upsert(COMPANY_ID, {
|
|
profileKey: "managed",
|
|
displayName: "Managed Agent",
|
|
anthropicAgentId: "agent-example",
|
|
agentVersion: "1",
|
|
environmentId: "environment-example",
|
|
defaultModel: "claude-sonnet-5",
|
|
defaultMaxListCostUsd: 2.5,
|
|
apiKeySecretId: rotatedSecretId,
|
|
enabled: true,
|
|
retentionAcknowledged: true,
|
|
qualification,
|
|
});
|
|
expect(updatedManaged).toMatchObject({
|
|
apiKeySecretId: rotatedSecretId,
|
|
defaultMaxListCostCents: 250,
|
|
qualifiedRevision: existingManaged.qualifiedRevision,
|
|
});
|
|
|
|
const awsQualification = { suite: "aws-agentcore-harness-context-v2" };
|
|
const existingRemote = {
|
|
id: "30000000-0000-4000-8000-000000000005",
|
|
companyId: COMPANY_ID,
|
|
profileKey: "agentcore",
|
|
displayName: "AgentCore",
|
|
service: "aws_bedrock_agentcore_harness",
|
|
configuration: { ...AWS_CONFIGURATION },
|
|
enabled: true,
|
|
retentionAcknowledged: true,
|
|
qualification: awsQualification,
|
|
qualifiedAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
qualifiedRevision: computeRemoteAgentProfileRevision({
|
|
service: "aws_bedrock_agentcore_harness",
|
|
configuration: AWS_CONFIGURATION,
|
|
retentionAcknowledged: true,
|
|
qualification: awsQualification,
|
|
}),
|
|
createdAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
updatedAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
};
|
|
const updatedRemote = await remoteAgentProfileService(
|
|
dbForUpsert(existingRemote),
|
|
).upsert(COMPANY_ID, remoteInput({
|
|
enabled: true,
|
|
retentionAcknowledged: true,
|
|
configuration: {
|
|
...AWS_CONFIGURATION,
|
|
defaultMaxEstimatedSessionCostUsd: 3,
|
|
},
|
|
}));
|
|
expect(updatedRemote).toMatchObject({
|
|
configuration: {
|
|
defaultModel: "global.anthropic.claude-sonnet-4-6",
|
|
defaultMaxEstimatedSessionCostUsd: 3,
|
|
},
|
|
qualifiedRevision: existingRemote.qualifiedRevision,
|
|
});
|
|
});
|
|
|
|
it("rejects runtime use when stored identity drifts from the qualified revision", async () => {
|
|
const qualification = { suite: "aws-agentcore-harness-context-v2" };
|
|
const qualifiedRevision = computeRemoteAgentProfileRevision({
|
|
service: "aws_bedrock_agentcore_harness",
|
|
configuration: AWS_CONFIGURATION,
|
|
retentionAcknowledged: true,
|
|
qualification,
|
|
});
|
|
await expect(
|
|
remoteAgentProfileService(dbReturningFirstRow({
|
|
id: "30000000-0000-4000-8000-000000000003",
|
|
companyId: COMPANY_ID,
|
|
profileKey: "agentcore",
|
|
displayName: "AgentCore",
|
|
service: "aws_bedrock_agentcore_harness",
|
|
configuration: { ...AWS_CONFIGURATION, memoryId: "tampered-memory" },
|
|
enabled: true,
|
|
retentionAcknowledged: true,
|
|
qualification,
|
|
qualifiedAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
qualifiedRevision,
|
|
createdAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
updatedAt: new Date("2026-08-01T00:00:00.000Z"),
|
|
})).requireQualified(COMPANY_ID, "agentcore"),
|
|
).rejects.toThrow("does not match its qualified revision");
|
|
});
|
|
});
|