Files
PaperClipAI/tests/runner-e2e/support.test.ts
T
Dotta 0f94521017 fix(runner): restore local session and task integrity (#12721)
## Thinking Path

> - Paperclip is the control plane for agents that perform work.
> - Paperclip Runner connects durable provider sessions to individual
task runs through PRP.
> - Provider continuity and per-run authority are different lifetimes.
> - The existing implementation mixed those lifetimes and lost event
metadata between provider frames, runnerd, persistence, API
sanitization, and the task thread.
> - That caused failed continuation, missing progress and Plans,
duplicate replies, hidden failures, and unsafe recovery.
> - This repair gives every heartbeat fresh authority, preserves
qualified provider-session continuity, and restores one lossless
presentation path without changing direct adapters.

## Linked Issues or Issue Description

**What happened?**

A second native heartbeat could reuse tickets, leases, command receipts,
sequence state, and run identity from the first heartbeat. Provider
phase and item identity could be lost before the UI read them. Redaction
could corrupt protocol discriminators while still missing malformed
credential tails. The task thread could fold progress into the final
response, hide failures, or show more than one final answer. Native
Codex also exposed approval modes that do not yet have a durable
approval bridge.

**Expected behavior**

Each heartbeat uses a new PRP authority epoch. Codex and OpenCode
preserve exact qualified provider sessions; ACPX emits an explicit
continuity event when its qualified process-replacement policy is used.
Every accepted provider event is presented, classified as internal, or
surfaced as unsupported. The task page shows chronological progress,
reasoning summaries, activity, Plans, interactions, terminal failures,
and exactly one final reply. Direct adapters retain their existing path.

**Steps to reproduce**

1. Enable the unified experimental Paperclip Runner setting.
2. Create a local native Codex, OpenCode, ACPX Claude, or ACPX Codex
agent.
3. Run response, Plan, structured-question/resume, restart,
cancellation, and failure scenarios.
4. Reload the task while active, waiting, failed, and settled.
5. On the old implementation, observe stale run authority, missing
classifications, incomplete output, or duplicated/folded replies.

**Paperclip version or commit**

The repair is based directly on `master` at
`87d05e194b643810d16d20612115acd01d735d43`.

**Deployment mode**

Local development with the embedded database.

Related work: Refs #12616, #12646, #12666, #12685, and #12700.

## What Changed

- Rotates PRP control-plane, outbox, ticket, lease, command, receipt,
and sequence authority for each heartbeat while carrying forward only a
validated provider-session identity.
- Reads `control-plane-state.json`, validates both durable schemas and
lifecycle values, resumes coherent current runs, archives qualified
settled authority, and quarantines malformed or mismatched scoped state
without moving ambiguous live legacy state.
- Preserves Codex provider phase and stable item identities so
commentary remains progress and only `final_answer` becomes final.
- Adds raw OpenCode HTTP/SSE boundary coverage and canonical reasoning
lifecycle mapping.
- Makes ACPX normalization lossless for visible reasoning, tool
lifecycle metadata, stable bounded identities, Plan revisions,
structured requests, failures, and qualified process replacement. Only
the compatible terminal assistant message is promoted as final.
- Applies schema-aware redaction before generic JWT-shaped detection and
scans every diagnostic string leaf. Malformed raw/escaped quoted
credential tails are redacted in both server and durable Rust state.
- Restores snapshot-style chronological task presentation, expandable
tool activity, inline Plan cards, visible waiting/resume/cancel/failure
states, and exactly one final answer.
- Makes `never` the only qualified native Codex permission mode and
rejects unsupported persisted native modes with remediation. OpenCode
and ACPX policies remain intact.
- Keeps the unified experimental Runner setting as the only enablement
flag. Onboarding and direct Codex, Claude, and OpenCode stay on their
legacy execution/finalization paths.
- Adds cross-language goldens, authority/recovery/fault coverage, exact
response/count assertions, and native plus legacy acceptance scenarios.

## Verification

- Pull-request GitHub Actions run Rust formatting/tests, TypeScript
checks, server/UI tests, builds, protocol drift checks, browser E2E, and
security scans.
- A separate workflow-only validation ref is pinned directly on this PR
head and runs the 35-cell paid local matrix: three core scenarios plus
structured-question resume and restart/resume for native Codex, native
OpenCode, ACPX Claude, ACPX Codex, and direct Codex/Claude/OpenCode.
Run: https://github.com/paperclipai/paperclip/actions/runs/33682434315
- Acceptance requires exact single visible replies, monotonic sequences,
matching envelope discriminators, one semantic terminal, one run
terminal, no unresolved interaction, no duplicate mutation, no secret
leakage, provider continuity, and zero native rows for direct adapters.
- Per maintainer direction, tests are running in GitHub Actions rather
than on the slower local host. Only formatters and static diff checks
were run locally.

## Risks

- Recovery from old or partial filesystem state is sensitive. The repair
fails closed, preserves active or unverifiable authority, and
quarantines only state whose scoped ownership is safe to move.
- Provider event formats can change. Closed validators and boundary
goldens turn new or malformed events into visible diagnostics instead of
silent drops.
- Shared task presentation could affect direct adapters. Runtime-fact
gating plus the direct-adapter matrix protect the existing path.
- Managed and remote providers are not qualified here. Shared code
continues to compile and fail safely, but live qualification is
deferred.

> 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 based on GPT-5. The exact deployed snapshot and
context-window size are not exposed to this task. It used agentic
reasoning, repository inspection, code editing, Git, parallel subagents,
and GitHub Actions.

## 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 and contains no internal
Paperclip ticket id
- [ ] I have run tests locally and they pass (intentionally deferred to
GitHub Actions)
- [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 risks above
- [ ] All Paperclip CI gates are green
- [ ] The paid local-provider matrix is 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
2026-09-02 16:11:26 -05:00

734 lines
23 KiB
TypeScript

import { execFileSync } from "node:child_process";
import { mkdir, mkdtemp, readFile, rm, writeFile } from "node:fs/promises";
import os from "node:os";
import path from "node:path";
import { afterEach, describe, expect, it, vi } from "vitest";
import { packageEvidence } from "./evidence.js";
import { RunnerApi } from "./api.js";
import { FixtureRegistry } from "./fixture-registry.js";
import { classifyFailure, shouldRetryFailure } from "./failure-classifier.js";
import {
assertIsolatedServerEnvironment,
buildPaperclipServerEnvironment,
resolvePaperclipRunnerBinaryForHarness,
} from "./harness-env.js";
import { runnerExecutionById } from "./catalog.js";
import { assertEmbeddedDatabaseIsolation } from "./instance-isolation.js";
import { evaluateMatchers } from "./matchers.js";
import {
assertSecretFree,
findSecretLeak,
findSecretLeakInJsonValues,
findSecretLeakInDirectory,
isEphemeralCodexRuntimeAuthFile,
redactText,
sanitizeJson,
} from "./redaction.js";
import { parseDarwinSharedMemory } from "./shared-memory.js";
import {
isNonExecutingReviewFenceRun,
numberedPlanStepCount,
} from "./run-observations.js";
import { runnerE2EWebServerCommand } from "./web-server-command.js";
const cleanupDirectories: string[] = [];
afterEach(async () => {
vi.unstubAllEnvs();
vi.unstubAllGlobals();
await Promise.all(
cleanupDirectories
.splice(0)
.map((directory) => rm(directory, { recursive: true, force: true })),
);
});
describe("runner E2E local binary resolution", () => {
const localNativeExecution = runnerExecutionById(
"core-compatibility.runner-codex.local.message-marker",
);
it("uses the debug runner binary built by the E2E workflow", () => {
expect(
resolvePaperclipRunnerBinaryForHarness(
[localNativeExecution],
"/repository",
undefined,
"linux",
),
).toBe(
path.join(
"/repository",
"packages/paperclip-runner/runner/target/debug/paperclip-runnerd",
),
);
});
it("preserves an explicit runner binary override", () => {
expect(
resolvePaperclipRunnerBinaryForHarness(
[localNativeExecution],
"/repository",
"/custom/paperclip-runnerd",
"linux",
),
).toBe("/custom/paperclip-runnerd");
});
});
describe("runner E2E sensitive API boundary", () => {
it("keeps secret request bodies out of Playwright API tracing", async () => {
vi.stubEnv("PAPERCLIP_RUNNER_E2E_PORT", "43123");
const playwrightPost = vi.fn();
const fetchMock = vi.fn().mockResolvedValue(
new Response(JSON.stringify({ id: "secret-id" }), {
status: 201,
headers: { "content-type": "application/json" },
}),
);
vi.stubGlobal("fetch", fetchMock);
const api = new RunnerApi({ post: playwrightPost } as never);
await expect(
api.postSensitive("/api/companies/company-id/secrets", {
value: "fixture-secret-value",
}),
).resolves.toEqual({ id: "secret-id" });
expect(playwrightPost).not.toHaveBeenCalled();
expect(fetchMock).toHaveBeenCalledWith(
new URL("http://127.0.0.1:43123/api/companies/company-id/secrets"),
expect.objectContaining({ method: "POST" }),
);
});
});
describe("runner E2E structured evidence scanning", () => {
it("does not invent a secret shape across JSON syntax boundaries", () => {
const value = {
OPENROUTER_API_KEY: "generated-secret-reference-id",
};
expect(findSecretLeak(JSON.stringify(value), [])).toBe(
"secret-shaped value",
);
expect(findSecretLeakInJsonValues(value, [])).toBeNull();
});
it("still rejects exact and provider-shaped values in nested JSON", () => {
expect(
findSecretLeakInJsonValues(
{ nested: ["safe", "campaign-secret-value"] },
["campaign-secret-value"],
),
).toBe("exact secret value");
expect(
findSecretLeakInJsonValues({ nested: "sk-proj-abcdefghijklmnop" }, []),
).toBe("secret-shaped value");
});
});
describe("runner E2E fixture registry", () => {
it("sets up in dependency order and tears down in reverse", async () => {
const events: string[] = [];
const registry = new FixtureRegistry()
.register({
id: "company",
setup: async () => {
events.push("setup-company");
return "c";
},
teardown: async () => {
events.push("teardown-company");
},
})
.register({
id: "agent",
dependencies: ["company"],
setup: async () => {
events.push("setup-agent");
return "a";
},
teardown: async () => {
events.push("teardown-agent");
},
});
const active = await registry.setupAll();
await active.teardown();
expect(events).toEqual([
"setup-company",
"setup-agent",
"teardown-agent",
"teardown-company",
]);
});
it("tears down partial setup after a failure", async () => {
const cleanup = vi.fn();
const registry = new FixtureRegistry()
.register({ id: "company", setup: async () => "c", teardown: cleanup })
.register({
id: "agent",
dependencies: ["company"],
setup: async () => {
throw new Error("boom");
},
});
await expect(registry.setupAll()).rejects.toThrow("boom");
expect(cleanup).toHaveBeenCalledOnce();
});
it("preserves setup and partial-cleanup failures", async () => {
const registry = new FixtureRegistry()
.register({
id: "company",
setup: async () => "c",
teardown: async () => {
throw new Error("cleanup transport returned 503");
},
})
.register({
id: "agent",
dependencies: ["company"],
setup: async () => {
throw new Error("agent setup failed");
},
});
await expect(registry.setupAll()).rejects.toMatchObject({
name: "AggregateError",
message: expect.stringContaining("cleanup failed"),
errors: expect.arrayContaining([
expect.objectContaining({ message: "agent setup failed" }),
expect.objectContaining({ message: "Fixture teardown failed" }),
]),
});
});
});
describe("runner E2E matchers", () => {
it("normalizes message text and evaluates state invariants", async () => {
const results = await evaluateMatchers(
[
{ kind: "message_contains", expected: "PAPERCLIP_E2E_OK_nonce" },
{
kind: "message_occurrences",
expected: "PAPERCLIP_E2E_OK_nonce",
count: 1,
},
{ kind: "issue_status", expected: "done" },
{ kind: "runtime_mode", expected: "native" },
],
{
message: " complete PAPERCLIP\\_E2E\\_OK\\_nonce ",
issueStatus: "done",
runtimeMode: "native",
},
);
expect(results.every((result) => result.passed)).toBe(true);
});
it("requires an exact number of complete visible markers", async () => {
const [result] = await evaluateMatchers(
[{ kind: "message_occurrences", expected: "FINAL_marker", count: 1 }],
{ message: "FINAL\\_marker\nFINAL_marker" },
);
expect(result).toMatchObject({ passed: false });
expect(result?.detail).toContain("observed 2");
});
it("normalizes ordered fragments and evaluates nested JSON Schema", async () => {
const results = await evaluateMatchers(
[
{
kind: "message_ordered",
expected: ["first marker", "second marker"],
},
{
kind: "json_schema",
schema: {
type: "object",
required: ["run"],
additionalProperties: false,
properties: {
run: {
type: "object",
required: ["status"],
properties: { status: { const: "succeeded" } },
},
},
},
},
],
{
message: "first marker\nsecond marker",
json: { run: { status: "succeeded" } },
},
);
expect(results.every((result) => result.passed)).toBe(true);
});
});
describe("runner E2E run observations", () => {
it("counts provider-equivalent numbered Plan step formats", () => {
expect(numberedPlanStepCount("1. First\n2) Second")).toBe(2);
expect(
numberedPlanStepCount(
"# Plan\n\nStep 1 — First\n\nStep 2 — Second\n\nStep 3: Verify",
),
).toBe(3);
expect(
numberedPlanStepCount("## **Step 1** — First\n- **2.** Second"),
).toBe(2);
});
it("excludes only queued continuations fenced while awaiting review", () => {
expect(
isNonExecutingReviewFenceRun({
status: "cancelled",
errorCode: "issue_continuation_waiting_on_review",
}),
).toBe(true);
expect(
isNonExecutingReviewFenceRun({
status: "failed",
errorCode: "issue_continuation_waiting_on_review",
}),
).toBe(false);
expect(
isNonExecutingReviewFenceRun({
status: "cancelled",
errorCode: "provider_failure",
}),
).toBe(false);
});
});
describe("runner E2E failure policy", () => {
it("retries only transient infrastructure failures", () => {
expect(
classifyFailure(new Error("Daytona preview connection timed out")),
).toBe("transient_infrastructure");
expect(
classifyFailure(
new Error(
"Browser bootstrap failed before task creation: New Task button timed out after a Vite 504",
),
),
).toBe("transient_infrastructure");
expect(
shouldRetryFailure(
classifyFailure(new Error("Daytona preview connection timed out")),
),
).toBe(true);
expect(
shouldRetryFailure(classifyFailure(new Error("marker matcher failed"))),
).toBe(false);
expect(
classifyFailure(
new Error(
"Timed out waiting for issue abc and heartbeat run terminal state",
),
),
).toBe("candidate_failure");
expect(
classifyFailure(
new Error("Provider request timed out during generation"),
),
).toBe("transient_infrastructure");
expect(
classifyFailure(
new Error(
"runner_ingress_unavailable: paperclip-runnerd: cumulative ACK cannot move beyond the produced source cursor",
),
),
).toBe("transient_infrastructure");
expect(
shouldRetryFailure(classifyFailure(new Error("invalid API key"))),
).toBe(false);
expect(
classifyFailure(
new Error("Daytona lease cleanup failed: provider returned 503"),
),
).toBe("transient_infrastructure");
expect(classifyFailure(new Error("cleanup invariant failure"))).toBe(
"cleanup_failure",
);
});
});
describe("runner E2E server isolation", () => {
it("strips database and paid-provider credentials from the Paperclip process", () => {
const env = buildPaperclipServerEnvironment(
{
PATH: "/bin",
DATABASE_URL: "postgres://existing",
DATABASE_MIGRATION_URL: "postgres://migration",
OPENAI_API_KEY: "openai",
ANTHROPIC_API_KEY: "anthropic",
OPENROUTER_API_KEY: "openrouter",
DAYTONA_API_KEY: "daytona",
OPENAI_ORG_ID: "also-provider-sensitive",
PAPERCLIP_API_KEY: "ambient-board-key",
PAPERCLIP_AGENT_API_KEY: "ambient-agent-key",
PAPERCLIP_TASK_BRIDGE_TOKEN: "ambient-task-token",
PAPERCLIP_SETUP_TOKEN: "ambient-setup-token",
PAPERCLIP_SECRETS_MASTER_KEY: "ambient-master-key",
PAPERCLIP_SECRETS_MASTER_KEY_FILE: "/outside/master.key",
PAPERCLIP_STORAGE_S3_BUCKET: "production-bucket",
},
{
PAPERCLIP_HOME: "/tmp/cell/paperclip-home",
PAPERCLIP_CONFIG: "/tmp/cell/paperclip-home/instances/e2e/config.json",
PAPERCLIP_AGENT_JWT_SECRET: "generated-agent-jwt",
PAPERCLIP_DECISION_SIGNING_SECRET: "generated-decision-key",
PAPERCLIP_TOOL_ACTION_SIGNING_SECRET: "generated-tool-key",
BETTER_AUTH_SECRET: "generated-auth-key",
},
);
expect(env.PATH).toBe("/bin");
expect(env.DATABASE_URL).toBeUndefined();
expect(env.OPENAI_API_KEY).toBeUndefined();
expect(env.OPENAI_ORG_ID).toBeUndefined();
expect(env.PAPERCLIP_API_KEY).toBeUndefined();
expect(env.PAPERCLIP_AGENT_API_KEY).toBeUndefined();
expect(env.PAPERCLIP_AGENT_JWT_SECRET).toBe("generated-agent-jwt");
expect(env.PAPERCLIP_TASK_BRIDGE_TOKEN).toBeUndefined();
expect(env.PAPERCLIP_SETUP_TOKEN).toBeUndefined();
expect(env.PAPERCLIP_SECRETS_MASTER_KEY).toBeUndefined();
expect(env.PAPERCLIP_SECRETS_MASTER_KEY_FILE).toBeUndefined();
expect(env.PAPERCLIP_STORAGE_S3_BUCKET).toBeUndefined();
expect(() =>
assertIsolatedServerEnvironment(env, {
temporaryRoot: "/tmp/cell",
paperclipHome: "/tmp/cell/paperclip-home",
configPath: "/tmp/cell/paperclip-home/instances/e2e/config.json",
}),
).not.toThrow();
});
it("uses absolute repository paths for the Playwright web server", () => {
const command = runnerE2EWebServerCommand("/workspace/paperclip");
expect(command).toContain(
"'/workspace/paperclip/cli/node_modules/tsx/dist/cli.mjs'",
);
expect(command).toContain(
"'/workspace/paperclip/tests/runner-e2e/server.ts'",
);
});
it("accepts only embedded state paths beneath the temporary root", async () => {
const root = await mkdtemp(
path.join(os.tmpdir(), "runner-e2e-instance-isolation-test-"),
);
cleanupDirectories.push(root);
const database = path.join(root, "paperclip-home", "db");
const secretsKey = path.join(
root,
"paperclip-home",
"secrets",
"master.key",
);
const configPath = path.join(root, "paperclip-home", "config.json");
await mkdir(database, { recursive: true });
await mkdir(path.dirname(secretsKey), { recursive: true });
await writeFile(secretsKey, "generated-master-key");
const config = {
database: {
mode: "embedded-postgres",
embeddedPostgresDataDir: database,
backup: { dir: path.join(root, "backups") },
},
logging: { logDir: path.join(root, "logs") },
storage: {
provider: "local_disk",
localDisk: { baseDir: path.join(root, "storage") },
},
secrets: {
provider: "local_encrypted",
strictMode: true,
localEncrypted: { keyFilePath: secretsKey },
},
};
await writeFile(configPath, JSON.stringify(config));
await expect(
assertEmbeddedDatabaseIsolation(configPath, root),
).resolves.toBeUndefined();
await writeFile(
configPath,
JSON.stringify({
...config,
storage: {
...config.storage,
localDisk: { baseDir: "/outside/storage" },
},
}),
);
await expect(
assertEmbeddedDatabaseIsolation(configPath, root),
).rejects.toThrow("storage path escaped");
});
});
describe("runner E2E evidence redaction", () => {
const secret = "sk-proj-supersecretvalue123456";
it("redacts exact and shaped credentials recursively", () => {
expect(redactText(`token=${secret}`, [secret])).toBe("token=[REDACTED]");
expect(sanitizeJson({ nested: [secret] }, [secret])).toEqual({
nested: ["[REDACTED]"],
});
expect(sanitizeJson("paperclip.runner-e2e.evidence/v1", [secret])).toBe(
"paperclip.runner-e2e.evidence/v1",
);
});
it("detects leaks and accepts sanitized evidence", () => {
expect(findSecretLeak(Buffer.from(secret), [secret])).toBeTruthy();
expect(() => assertSecretFree("safe", [secret], "fixture")).not.toThrow();
expect(() =>
assertSecretFree(
"sk-proj-documentationfixture123456",
[secret],
"API source",
{ includeShapes: false },
),
).not.toThrow();
});
it("finds exact credentials across streamed persisted-state chunks", async () => {
const root = await mkdtemp(
path.join(os.tmpdir(), "runner-e2e-secret-scan-test-"),
);
cleanupDirectories.push(root);
await writeFile(
path.join(root, "database.bin"),
Buffer.concat([
Buffer.alloc(65_530, "x"),
Buffer.from(secret),
Buffer.alloc(32, "y"),
]),
);
await expect(
findSecretLeakInDirectory(root, [secret]),
).resolves.toMatchObject({ reason: "exact secret value" });
});
it("can ignore fake key shapes while scanning persisted package state", async () => {
const root = await mkdtemp(
path.join(os.tmpdir(), "runner-e2e-secret-shape-test-"),
);
cleanupDirectories.push(root);
await writeFile(
path.join(root, "provider-fixture.test.ts"),
'const fake = "sk-proj-documentationfixture123456";\n',
);
await expect(
findSecretLeakInDirectory(root, [secret], { includeShapes: false }),
).resolves.toBeNull();
});
it("can narrowly exclude a verified ephemeral runtime credential file", async () => {
const root = await mkdtemp(
path.join(os.tmpdir(), "runner-e2e-ephemeral-auth-test-"),
);
cleanupDirectories.push(root);
const runtimeAuth = path.join(root, "codex-home", "auth.json");
const forbiddenConfig = path.join(root, "config.json");
await mkdir(path.dirname(runtimeAuth), { recursive: true });
await writeFile(runtimeAuth, secret, { mode: 0o600 });
await writeFile(forbiddenConfig, secret);
await expect(
findSecretLeakInDirectory(root, [secret], {
includeShapes: false,
ignoreFile: (file) => file === runtimeAuth,
}),
).resolves.toMatchObject({ file: forbiddenConfig });
});
it("recognizes both managed and durable-session Codex runtime auth files", () => {
const root = path.join(os.tmpdir(), "paperclip-home");
expect(
isEphemeralCodexRuntimeAuthFile(
root,
path.join(
root,
"instances/instance-1/companies/company-1/agents/agent-1/codex-home/auth.json",
),
),
).toBe(true);
expect(
isEphemeralCodexRuntimeAuthFile(
root,
path.join(
root,
"instances/instance-1/runtime/paperclip-runner/durable-sessions/session-1/codex-home/auth.json",
),
),
).toBe(true);
expect(
isEphemeralCodexRuntimeAuthFile(
root,
path.join(root, "instances/instance-1/runtime/auth.json"),
),
).toBe(false);
expect(
isEphemeralCodexRuntimeAuthFile(
root,
path.join(
root,
"instances/instance-1/runtime/paperclip-runner/durable-sessions/session-1/codex-home/config.toml",
),
),
).toBe(false);
});
it("publishes only allowlisted sanitized files and reports source leaks", async () => {
const root = await mkdtemp(
path.join(os.tmpdir(), "runner-e2e-evidence-test-"),
);
cleanupDirectories.push(root);
const privateDir = path.join(root, "private");
const uploadDir = path.join(root, "upload");
await mkdir(privateDir, { recursive: true });
await writeFile(
path.join(privateDir, "result.json"),
JSON.stringify({ error: secret }),
);
await writeFile(path.join(privateDir, "database.sqlite"), secret);
const packaged = await packageEvidence({
privateDir,
uploadDir,
secrets: [secret],
expectPassScreenshot: false,
});
expect(packaged.leaks).toEqual([
{ file: "result.json", reason: "exact secret value" },
]);
expect(
await readFile(path.join(uploadDir, "result.json"), "utf8"),
).toContain("[REDACTED]");
await expect(
readFile(path.join(uploadDir, "database.sqlite")),
).rejects.toThrow();
});
it("keeps raster evidence private to CI and rejects active SVG content", async () => {
const root = await mkdtemp(
path.join(os.tmpdir(), "runner-e2e-visual-evidence-test-"),
);
cleanupDirectories.push(root);
const privateDir = path.join(root, "private");
const uploadDir = path.join(root, "upload");
const playwrightOutput = path.join(privateDir, "playwright-output");
await mkdir(playwrightOutput, { recursive: true });
await writeFile(path.join(privateDir, "final-state.png"), "png");
await writeFile(path.join(playwrightOutput, "failure.webm"), "webm");
await writeFile(
path.join(playwrightOutput, "active.svg"),
"<svg onload='alert(1)' />",
);
const packaged = await packageEvidence({
privateDir,
uploadDir,
secrets: [secret],
expectPassScreenshot: false,
});
expect(packaged.files).toEqual(
expect.arrayContaining([
"final-state.png",
path.join("playwright-output", "failure.webm"),
]),
);
expect(packaged.files).not.toContain(
path.join("playwright-output", "active.svg"),
);
await expect(
readFile(path.join(playwrightOutput, "active.svg"), "utf8"),
).resolves.toContain("onload");
await expect(
readFile(path.join(uploadDir, "playwright-output", "active.svg")),
).rejects.toThrow();
});
it("preserves valid JSON while redacting escaped command diagnostics", async () => {
const root = await mkdtemp(
path.join(os.tmpdir(), "runner-e2e-json-evidence-test-"),
);
cleanupDirectories.push(root);
const privateDir = path.join(root, "private");
const uploadDir = path.join(root, "upload");
await mkdir(path.join(privateDir, "snapshots"), { recursive: true });
await writeFile(
path.join(privateDir, "snapshots", "api-state.json"),
JSON.stringify({
log: String.raw`curl -H \"Authorization: Bearer temporary-run-token\" \\\n+ \"$PAPERCLIP_API_URL/api/issues\"`,
}),
);
await packageEvidence({
privateDir,
uploadDir,
secrets: [secret],
expectPassScreenshot: false,
});
const uploaded = await readFile(
path.join(uploadDir, "snapshots", "api-state.json"),
"utf8",
);
expect(() => JSON.parse(uploaded)).not.toThrow();
expect(uploaded).toContain("[REDACTED]");
});
it("streams large ZIP evidence and detects exact secrets", async () => {
const root = await mkdtemp(
path.join(os.tmpdir(), "runner-e2e-zip-evidence-test-"),
);
cleanupDirectories.push(root);
const privateDir = path.join(root, "private");
const uploadDir = path.join(root, "upload");
const blobDir = path.join(privateDir, "blob-report");
await mkdir(blobDir, { recursive: true });
await writeFile(
path.join(blobDir, "trace.txt"),
Buffer.concat([Buffer.alloc(2 * 1024 * 1024, "x"), Buffer.from(secret)]),
);
execFileSync("zip", ["-q", "report.zip", "trace.txt"], {
cwd: blobDir,
});
await rm(path.join(blobDir, "trace.txt"));
const packaged = await packageEvidence({
privateDir,
uploadDir,
secrets: [secret],
expectPassScreenshot: false,
});
expect(packaged.leaks).toEqual([
{
file: path.join("blob-report", "report.zip"),
reason: "exact secret value",
},
]);
await expect(
readFile(path.join(uploadDir, "blob-report", "report.zip")),
).rejects.toThrow();
});
});
describe("runner E2E macOS shared-memory cleanup", () => {
it("parses only shared-memory rows from ipcs output", () => {
expect(
parseDarwinSharedMemory(
`IPC status from <running system>\nT ID KEY MODE OWNER GROUP CREATOR CGROUP NATTCH SEGSZ CPID LPID\nm 327709 0x028ed0ab --rw------- dotta staff dotta staff 0 56 52172 52172\ns 123 0x0 --ra------- dotta staff\n`,
),
).toEqual([
{
id: "327709",
owner: "dotta",
attachments: 0,
creatorPid: 52172,
},
]);
});
});