mirrors/pikasTech-agentrun
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: 收敛 codex stdio trace 噪声 (#66)

Browse Source 
Co-authored-by: Codex <codex@pikas.tech>
This commit is contained in:
Lyon
2026-06-02 09:49:55 +08:00
committed by GitHub
parent 5db48b299e
commit 1d45d272f1
4 changed files with 109 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/selftest/cases/30-codex-stdio.ts
+25 -1
View File
@@ -120,6 +120,25 @@ const selfTest: SelfTestCase = async (context) => {
const liveResult = await livePromise;
assert.equal(liveResult.terminalStatus, "completed", "slow live tool event turn should complete");
const noisy = await createRunWithCommand(client, context, "hello noisy reasoning", "selftest-noisy-reasoning-events", 15_000);
const noisyResult = await runOnce({ managerUrl: server.baseUrl, runId: noisy.runId, codexCommand: context.fakeCodexCommand, codexArgs: context.fakeCodexArgs, codexHome: context.codexHome, env: { CODEX_HOME: context.codexHome, AGENTRUN_FAKE_CODEX_MODE: "noisy-reasoning-events" }, oneShot: true }) as JsonRecord;
assert.equal(noisyResult.terminalStatus, "completed", "noisy reasoning turn should complete");
const noisyEvents = await client.get(`/api/v1/runs/${noisy.runId}/events?afterSeq=0&limit=100`) as { items?: Array<{ type: string; payload: unknown }> };
const noisyItems = noisyEvents.items ?? [];
assert.equal(noisyItems.some((event) => event.type === "backend_status" && eventPayload(event).phase === "item/reasoning/textDelta"), false, "reasoning textDelta must not be persisted as backend_status");
assert.equal(noisyItems.some((event) => event.type === "backend_status" && eventPayload(event).phase === "thread/tokenUsage/updated"), false, "token usage update must not be persisted as backend_status");
assert.equal(noisyItems.some((event) => event.type === "backend_status" && eventPayload(event).phase === "account/rateLimits/updated"), false, "rate limit update must not be persisted as backend_status");
assert.equal(noisyItems.some((event) => event.type === "backend_status" && eventPayload(event).phase === "warning"), false, "low value warnings must not be persisted as backend_status");
assert.equal(noisyItems.some((event) => event.type === "backend_status" && eventPayload(event).phase === "configWarning"), false, "low value config warnings must not be persisted as backend_status");
assert.equal(noisyItems.some((event) => event.type === "tool_call" && eventPayloadItem(event).type === "reasoning"), false, "reasoning items must not be persisted as tool_call");
assert.ok(noisyItems.some((event) => event.type === "tool_call" && eventPayload(event).method === "item/started" && eventPayloadItem(event).type === "commandExecution"), "real commandExecution tool call should remain visible");
assert.ok(noisyItems.some((event) => event.type === "assistant_message" && eventPayload(event).text === "noise filtered final"), "final assistant_message should remain visible");
const suppression = noisyItems.find((event) => event.type === "backend_status" && eventPayload(event).phase === "codex-app-server-notifications-suppressed");
assert.ok(suppression, "suppression summary must be emitted when noisy notifications are filtered");
assert.equal(eventPayload(suppression ?? { payload: {} }).total, 8);
assert.equal(JSON.stringify(noisyEvents).includes("internal reasoning must not become durable trace text"), false, "suppression summary must not leak reasoning text");
assertNoSecretLeak(noisyEvents);
await runFailureCase({ client, managerUrl: server.baseUrl, context, mode: "missing-turn-result", expectedStatus: "failed", expectedFailureKind: "backend-response-invalid" });
await runFailureCase({ client, managerUrl: server.baseUrl, context, mode: "provider-401-rpc-error", expectedStatus: "failed", expectedFailureKind: "provider-auth-failed" });
await runFailureCase({ client, managerUrl: server.baseUrl, context, mode: "provider-429-terminal", expectedStatus: "failed", expectedFailureKind: "provider-rate-limited" });
@@ -131,7 +150,7 @@ const selfTest: SelfTestCase = async (context) => {
await runSecretFailureCase({ client, managerUrl: server.baseUrl, context });
await runSpawnFailureCase({ client, managerUrl: server.baseUrl, context });
return { name: "codex-stdio", tests: ["runner-lease-heartbeat", "codex-stdio-fake-turn", "codex-stdio-projected-writable-home", "codex-stdio-deepseek-profile-fake-turn", "codex-stdio-minimax-m3-profile-fake-turn", "codex-stdio-deepseek-missing-secret-no-fallback", "codex-stdio-minimax-m3-missing-secret-no-fallback", "codex-stdio-config-model-authoritative", "codex-stdio-explicit-model-forwarded", "codex-stdio-final-agent-message-only", "codex-stdio-stale-thread-fallback", "codex-stdio-live-tool-events", "codex-stdio-missing-turn-result", "codex-stdio-provider-auth-failed", "codex-stdio-provider-rate-limited", "codex-stdio-provider-503-rpc-error", "codex-stdio-provider-503-terminal", "codex-stdio-provider-503-retry-event", "codex-stdio-invalid-json", "codex-stdio-timeout", "codex-stdio-secret-unavailable", "codex-stdio-spawn-failure"] };
return { name: "codex-stdio", tests: ["runner-lease-heartbeat", "codex-stdio-fake-turn", "codex-stdio-projected-writable-home", "codex-stdio-deepseek-profile-fake-turn", "codex-stdio-minimax-m3-profile-fake-turn", "codex-stdio-deepseek-missing-secret-no-fallback", "codex-stdio-minimax-m3-missing-secret-no-fallback", "codex-stdio-config-model-authoritative", "codex-stdio-explicit-model-forwarded", "codex-stdio-final-agent-message-only", "codex-stdio-stale-thread-fallback", "codex-stdio-live-tool-events", "codex-stdio-noisy-reasoning-suppression", "codex-stdio-missing-turn-result", "codex-stdio-provider-auth-failed", "codex-stdio-provider-rate-limited", "codex-stdio-provider-503-rpc-error", "codex-stdio-provider-503-terminal", "codex-stdio-provider-503-retry-event", "codex-stdio-invalid-json", "codex-stdio-timeout", "codex-stdio-secret-unavailable", "codex-stdio-spawn-failure"] };
} finally {
await new Promise<void>((resolve) => server.server.close(() => resolve()));
}
@@ -168,6 +187,11 @@ function eventPayload(event: { payload: unknown }): JsonRecord {
return typeof event.payload === "object" && event.payload !== null && !Array.isArray(event.payload) ? event.payload as JsonRecord : {};
}
function eventPayloadItem(event: { payload: unknown }): JsonRecord {
const item = eventPayload(event).item;
return typeof item === "object" && item !== null && !Array.isArray(item) ? item as JsonRecord : {};
}
async function waitForEvent(client: ManagerClient, runId: string, predicate: (event: { type: string; payload: unknown }) => boolean, label: string): Promise<void> {
const deadline = Date.now() + 3_000;
while (Date.now() < deadline) {
src/selftest/fake-codex-app-server.ts
+19
View File
@@ -148,6 +148,25 @@ for await (const line of rl) {
}, 50);
continue;
}
if (mode === "noisy-reasoning-events") {
turnCounter += 1;
const turn = { id: `turn_selftest_${turnCounter}`, status: "completed" };
notify("turn/started", { turn });
notify("item/started", { item: { id: "reasoning_selftest", type: "reasoning", content: [] } });
notify("item/reasoning/textDelta", { itemId: "reasoning_selftest", delta: "internal reasoning must not become durable trace text 1" });
notify("item/reasoning/textDelta", { itemId: "reasoning_selftest", delta: "internal reasoning must not become durable trace text 2" });
notify("thread/tokenUsage/updated", { usage: { inputTokens: 1, outputTokens: 2 } });
notify("account/rateLimits/updated", { limit: "selftest" });
notify("warning", { message: "low value warning should be summarized" });
notify("configWarning", { message: "low value config warning should be summarized" });
notify("item/completed", { item: { id: "reasoning_selftest", type: "reasoning", content: ["internal reasoning must not become durable trace text 3"] } });
notify("item/started", { item: { id: "tool_after_noise", type: "commandExecution", command: "echo after-noise" } });
notify("item/completed", { item: { id: "tool_after_noise", type: "commandExecution", command: "echo after-noise", status: "completed" } });
notify("item/completed", { item: { id: "msg_after_noise", type: "agentMessage", text: "noise filtered final" } });
notify("turn/completed", { turn });
respond(message.id, { turn });
continue;
}
turnCounter += 1;
const turn = { id: `turn_selftest_${turnCounter}`, status: "completed" };
notify("turn/started", { turn });