engineering-behavior-observatory 0.2.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/LICENSE +202 -0
- package/README.md +151 -0
- package/contracts/codex-app-server-0.150.1/manifest.json +25 -0
- package/contracts/codex-app-server-0.150.1/schema/InitializeParams.json +84 -0
- package/contracts/codex-app-server-0.150.1/schema/JSONRPCMessage.json +137 -0
- package/contracts/codex-app-server-0.150.1/schema/v2/ThreadReadParams.json +17 -0
- package/contracts/codex-app-server-0.150.1/schema/v2/ThreadTokenUsageUpdatedNotification.json +82 -0
- package/contracts/codex-app-server-0.150.1/schema/v2/TurnInterruptParams.json +17 -0
- package/contracts/codex-app-server-0.150.1/types/AskForApproval.ts +5 -0
- package/contracts/codex-app-server-0.150.1/types/ClientNotification.ts +5 -0
- package/contracts/codex-app-server-0.150.1/types/SandboxMode.ts +5 -0
- package/contracts/codex-app-server-0.150.1/types/ThreadReadParams.ts +9 -0
- package/contracts/codex-app-server-0.150.1/types/TokenUsageBreakdown.ts +5 -0
- package/contracts/codex-app-server-0.150.1/types/TurnInterruptParams.ts +5 -0
- package/contracts/codex-app-server-0.153.4/manifest.json +27 -0
- package/contracts/codex-app-server-0.153.4/schema/JSONRPCMessage.json +137 -0
- package/contracts/codex-app-server-0.153.4/schema/v1/InitializeParams.json +84 -0
- package/contracts/codex-app-server-0.153.4/schema/v2/ThreadReadParams.json +17 -0
- package/contracts/codex-app-server-0.153.4/schema/v2/ThreadStartParams.json +515 -0
- package/contracts/codex-app-server-0.153.4/schema/v2/ThreadTokenUsageUpdatedNotification.json +82 -0
- package/contracts/codex-app-server-0.153.4/schema/v2/TurnInterruptParams.json +17 -0
- package/contracts/codex-app-server-0.153.4/schema/v2/TurnStartParams.json +921 -0
- package/contracts/codex-app-server-0.153.4/types/AskForApproval.ts +5 -0
- package/contracts/codex-app-server-0.153.4/types/ClientNotification.ts +5 -0
- package/contracts/codex-app-server-0.153.4/types/SandboxMode.ts +5 -0
- package/contracts/codex-app-server-0.153.4/types/ThreadReadParams.ts +12 -0
- package/contracts/codex-app-server-0.153.4/types/TokenUsageBreakdown.ts +5 -0
- package/contracts/codex-app-server-0.153.4/types/TurnInterruptParams.ts +5 -0
- package/contracts/openhands-agent-server-v1.44.1.json +65 -0
- package/contracts/openhands-agent-server-v1.46.0.json +46 -0
- package/dist/contracts/codex-app-server-0.153.4/types/AskForApproval.d.ts +9 -0
- package/dist/contracts/codex-app-server-0.153.4/types/AskForApproval.js +2 -0
- package/dist/contracts/codex-app-server-0.153.4/types/ClientNotification.d.ts +3 -0
- package/dist/contracts/codex-app-server-0.153.4/types/ClientNotification.js +2 -0
- package/dist/contracts/codex-app-server-0.153.4/types/SandboxMode.d.ts +1 -0
- package/dist/contracts/codex-app-server-0.153.4/types/SandboxMode.js +2 -0
- package/dist/contracts/codex-app-server-0.153.4/types/ThreadReadParams.d.ts +10 -0
- package/dist/contracts/codex-app-server-0.153.4/types/ThreadReadParams.js +2 -0
- package/dist/contracts/codex-app-server-0.153.4/types/TokenUsageBreakdown.d.ts +8 -0
- package/dist/contracts/codex-app-server-0.153.4/types/TokenUsageBreakdown.js +2 -0
- package/dist/contracts/codex-app-server-0.153.4/types/TurnInterruptParams.d.ts +4 -0
- package/dist/contracts/codex-app-server-0.153.4/types/TurnInterruptParams.js +2 -0
- package/dist/src/agent-sdk-normalizer.d.ts +22 -0
- package/dist/src/agent-sdk-normalizer.js +995 -0
- package/dist/src/agent-sdk-run.d.ts +31 -0
- package/dist/src/agent-sdk-run.js +242 -0
- package/dist/src/agent-sdk-runner.d.ts +96 -0
- package/dist/src/agent-sdk-runner.js +407 -0
- package/dist/src/agent-sdk.d.ts +294 -0
- package/dist/src/agent-sdk.js +644 -0
- package/dist/src/aggregation.d.ts +183 -0
- package/dist/src/aggregation.js +513 -0
- package/dist/src/artifacts.d.ts +36 -0
- package/dist/src/artifacts.js +1372 -0
- package/dist/src/atlas-grafana.d.ts +377 -0
- package/dist/src/atlas-grafana.js +47 -0
- package/dist/src/atlas-html.d.ts +2 -0
- package/dist/src/atlas-html.js +34 -0
- package/dist/src/atlas.d.ts +130 -0
- package/dist/src/atlas.js +235 -0
- package/dist/src/behavior-assertions.d.ts +95 -0
- package/dist/src/behavior-assertions.js +154 -0
- package/dist/src/cli.d.ts +5 -0
- package/dist/src/cli.js +957 -0
- package/dist/src/codex-judge.d.ts +44 -0
- package/dist/src/codex-judge.js +214 -0
- package/dist/src/codex-run.d.ts +114 -0
- package/dist/src/codex-run.js +519 -0
- package/dist/src/codex.d.ts +187 -0
- package/dist/src/codex.js +1028 -0
- package/dist/src/contracts.d.ts +155 -0
- package/dist/src/contracts.js +866 -0
- package/dist/src/corpus.d.ts +67 -0
- package/dist/src/corpus.js +540 -0
- package/dist/src/cursor-sdk-runner.d.ts +81 -0
- package/dist/src/cursor-sdk-runner.js +362 -0
- package/dist/src/cursor-sdk.d.ts +81 -0
- package/dist/src/cursor-sdk.js +1078 -0
- package/dist/src/deepseek-adapter.d.ts +152 -0
- package/dist/src/deepseek-adapter.js +777 -0
- package/dist/src/exports.d.ts +69 -0
- package/dist/src/exports.js +966 -0
- package/dist/src/human-calibration.d.ts +167 -0
- package/dist/src/human-calibration.js +618 -0
- package/dist/src/index.d.ts +71 -0
- package/dist/src/index.js +38 -0
- package/dist/src/lifecycle.d.ts +196 -0
- package/dist/src/lifecycle.js +1889 -0
- package/dist/src/normalization-integrity.d.ts +128 -0
- package/dist/src/normalization-integrity.js +429 -0
- package/dist/src/openhands-run.d.ts +36 -0
- package/dist/src/openhands-run.js +415 -0
- package/dist/src/openhands.d.ts +130 -0
- package/dist/src/openhands.js +859 -0
- package/dist/src/pi.d.ts +203 -0
- package/dist/src/pi.js +1345 -0
- package/dist/src/process-protocol.d.ts +245 -0
- package/dist/src/process-protocol.js +1261 -0
- package/dist/src/retained-evidence.d.ts +12 -0
- package/dist/src/retained-evidence.js +201 -0
- package/dist/src/run-bundles.d.ts +193 -0
- package/dist/src/run-bundles.js +993 -0
- package/dist/src/scheduler.d.ts +109 -0
- package/dist/src/scheduler.js +1080 -0
- package/dist/src/semantic-judge.d.ts +194 -0
- package/dist/src/semantic-judge.js +875 -0
- package/dist/src/structural-observations.d.ts +124 -0
- package/dist/src/structural-observations.js +671 -0
- package/dist/src/task-packets.d.ts +115 -0
- package/dist/src/task-packets.js +683 -0
- package/dist/src/uniform-events.d.ts +120 -0
- package/dist/src/uniform-events.js +158 -0
- package/dist/src/verifiers.d.ts +116 -0
- package/dist/src/verifiers.js +819 -0
- package/dist/src/workspaces.d.ts +56 -0
- package/dist/src/workspaces.js +1314 -0
- package/docs/README.md +60 -0
- package/docs/development/README.md +71 -0
- package/docs/development/documentation-sync.md +58 -0
- package/docs/development/extension-contracts.md +307 -0
- package/docs/evaluation/README.md +48 -0
- package/docs/evaluation/aggregation.md +112 -0
- package/docs/evaluation/behavior-assertions.md +45 -0
- package/docs/evaluation/human-calibration.md +170 -0
- package/docs/evaluation/normalization-integrity.md +80 -0
- package/docs/evaluation/semantic-judge.md +188 -0
- package/docs/evaluation/structural-observations.md +84 -0
- package/docs/evaluation/uniform-events.md +131 -0
- package/docs/guides/atlas.md +144 -0
- package/docs/guides/evidence-and-sharing.md +100 -0
- package/docs/guides/operator-guide.md +424 -0
- package/docs/guides/quickstart.md +152 -0
- package/docs/guides/telemetry.md +66 -0
- package/docs/harnesses/README.md +31 -0
- package/docs/harnesses/claude-agent-sdk.md +61 -0
- package/docs/harnesses/codex-harness.md +174 -0
- package/docs/harnesses/cursor-sdk.md +149 -0
- package/docs/harnesses/deepseek-harness.md +134 -0
- package/docs/harnesses/openhands-agent-server.md +153 -0
- package/docs/harnesses/pi-sdk.md +66 -0
- package/docs/reference/README.md +19 -0
- package/docs/reference/agent-sdk-operational-runner.md +118 -0
- package/docs/reference/cli.md +114 -0
- package/docs/reference/contracts.md +222 -0
- package/docs/reference/run-bundle-contract.md +354 -0
- package/docs/reference/run-lifecycle.md +54 -0
- package/examples/cursor-sdk/README.md +11 -0
- package/examples/cursor-sdk/capture-profile.json +1 -0
- package/examples/cursor-sdk/harness.json +1 -0
- package/examples/cursor-sdk/model.json +1 -0
- package/examples/cursor-sdk/native-limits.json +1 -0
- package/examples/cursor-sdk/native-tool-policy.json +1 -0
- package/ontology/behavior-categories.v1.json +46 -0
- package/package.json +66 -0
- package/release/0.1.0/KNOWN_LIMITATIONS.md +35 -0
- package/release/0.1.0/README.md +46 -0
- package/release/0.1.0/reproducibility.json +61 -0
- package/release/0.2.0/KNOWN_LIMITATIONS.md +40 -0
- package/release/0.2.0/README.md +42 -0
- package/release/0.2.0/reproducibility.json +73 -0
- package/release/0.2.1/KNOWN_LIMITATIONS.md +18 -0
- package/release/0.2.1/README.md +63 -0
- package/release/0.2.1/reproducibility.json +73 -0
- package/release/README.md +13 -0
- package/schemas/aggregation.v1.json +293 -0
- package/schemas/behavior-assertions.v1.json +220 -0
- package/schemas/deepseek-runtime-composition.v1.schema.json +122 -0
- package/schemas/experiment.v1.schema.json +188 -0
- package/schemas/human-calibration.v1.json +336 -0
- package/schemas/normalization-integrity.v1.json +302 -0
- package/schemas/run-bundles/v1.json +1040 -0
- package/schemas/run-queue.v1.schema.json +206 -0
- package/schemas/semantic-judge.v1.json +380 -0
- package/schemas/structural-observations.v1.json +131 -0
- package/schemas/task-packet-freeze.v1.schema.json +106 -0
- package/schemas/task-packet.v1.schema.json +234 -0
- package/schemas/uniform-events/v1.json +289 -0
- package/scripts/atlas-grafana.sh +18 -0
|
@@ -0,0 +1,109 @@
|
|
|
1
|
+
import { type ArtifactValidationError } from "./artifacts.js";
|
|
2
|
+
import { type ArtifactReference, type DeclaredOrder, type DeclaredMatrixCell, type Digest, type ExperimentConfiguration, type ResolvedTaskPacket } from "./contracts.js";
|
|
3
|
+
import { type TaskPacketFreezeRecord } from "./task-packets.js";
|
|
4
|
+
export type QueueOrderingStrategy = "sequential" | "seeded-shuffle" | "balanced";
|
|
5
|
+
export type PermutationAlgorithm = "fisher-yates-v1";
|
|
6
|
+
export declare const MAX_RUN_QUEUE_ENTRIES = 100000;
|
|
7
|
+
export declare const MAX_RUN_QUEUE_BYTES: number;
|
|
8
|
+
export type FrozenTaskIdentity = {
|
|
9
|
+
id: string;
|
|
10
|
+
packetId: string;
|
|
11
|
+
packetRef: ArtifactReference;
|
|
12
|
+
freezeLocator: string;
|
|
13
|
+
aggregateDigest: Digest;
|
|
14
|
+
};
|
|
15
|
+
export type FrozenTaskInput = TaskPacketFreezeRecord | {
|
|
16
|
+
status?: "frozen";
|
|
17
|
+
packetId: string;
|
|
18
|
+
packetLocator?: string;
|
|
19
|
+
packetDigest: Digest;
|
|
20
|
+
freezeLocator: string;
|
|
21
|
+
aggregateDigest: Digest;
|
|
22
|
+
} | {
|
|
23
|
+
status?: "frozen";
|
|
24
|
+
packetId: string;
|
|
25
|
+
packetRef: ArtifactReference;
|
|
26
|
+
freezeLocator: string;
|
|
27
|
+
aggregateDigest: Digest;
|
|
28
|
+
};
|
|
29
|
+
export type RunQueueEntry = DeclaredMatrixCell & {
|
|
30
|
+
runId: string;
|
|
31
|
+
task: FrozenTaskIdentity;
|
|
32
|
+
model: {
|
|
33
|
+
id: string;
|
|
34
|
+
configurationRef: ArtifactReference;
|
|
35
|
+
};
|
|
36
|
+
harness: {
|
|
37
|
+
id: string;
|
|
38
|
+
configurationRef: ArtifactReference;
|
|
39
|
+
nativeLimitsRef: ArtifactReference;
|
|
40
|
+
nativeToolPolicyRef: ArtifactReference;
|
|
41
|
+
};
|
|
42
|
+
configuration: {
|
|
43
|
+
model: ArtifactReference;
|
|
44
|
+
harness: ArtifactReference;
|
|
45
|
+
nativeLimits: ArtifactReference;
|
|
46
|
+
nativeToolPolicy: ArtifactReference;
|
|
47
|
+
};
|
|
48
|
+
trial: {
|
|
49
|
+
index: number;
|
|
50
|
+
};
|
|
51
|
+
};
|
|
52
|
+
export type RunQueueMatrix = DeclaredOrder & {
|
|
53
|
+
trialCount: number;
|
|
54
|
+
};
|
|
55
|
+
export type RunQueue = {
|
|
56
|
+
schemaVersion: "ebo.run-queue/v1";
|
|
57
|
+
experimentId: string;
|
|
58
|
+
experimentDigest: Digest;
|
|
59
|
+
schedulingDigest: Digest;
|
|
60
|
+
captureProfile: ArtifactReference;
|
|
61
|
+
coordinatorBudget: {
|
|
62
|
+
maxWallClockMs: number;
|
|
63
|
+
};
|
|
64
|
+
matrix: RunQueueMatrix;
|
|
65
|
+
seed: string;
|
|
66
|
+
ordering: {
|
|
67
|
+
strategy: QueueOrderingStrategy;
|
|
68
|
+
balanceBy?: "task" | "model" | "harness";
|
|
69
|
+
permutationAlgorithm?: PermutationAlgorithm;
|
|
70
|
+
permutationAlgorithmRef?: ArtifactReference;
|
|
71
|
+
};
|
|
72
|
+
entries: RunQueueEntry[];
|
|
73
|
+
};
|
|
74
|
+
export type CompileRunQueueOptions = {
|
|
75
|
+
bundleRoot?: string;
|
|
76
|
+
resolvedDigests?: Record<string, Digest>;
|
|
77
|
+
permutationAlgorithm?: unknown;
|
|
78
|
+
permutationAlgorithms?: Record<string, unknown>;
|
|
79
|
+
freezeLocators?: Record<string, string>;
|
|
80
|
+
frozenTasks?: Record<string, FrozenTaskInput>;
|
|
81
|
+
taskFreezeRecords?: Record<string, FrozenTaskInput>;
|
|
82
|
+
taskPackets?: Record<string, FrozenTaskInput>;
|
|
83
|
+
resolvedPackets?: Record<string, ResolvedTaskPacket>;
|
|
84
|
+
};
|
|
85
|
+
export type ValidateRunQueueOptions = Pick<CompileRunQueueOptions, "bundleRoot" | "resolvedDigests" | "frozenTasks" | "taskFreezeRecords" | "taskPackets" | "permutationAlgorithm" | "permutationAlgorithms" | "freezeLocators" | "resolvedPackets">;
|
|
86
|
+
export type RunQueueInspection = {
|
|
87
|
+
experimentId: string;
|
|
88
|
+
seed: string;
|
|
89
|
+
strategy: QueueOrderingStrategy;
|
|
90
|
+
entryCount: number;
|
|
91
|
+
firstRunId: string | null;
|
|
92
|
+
lastRunId: string | null;
|
|
93
|
+
};
|
|
94
|
+
export declare function expandMatrixCells(experiment: ExperimentConfiguration): DeclaredMatrixCell[];
|
|
95
|
+
export declare function compileRunQueue(experiment: ExperimentConfiguration, options?: CompileRunQueueOptions): RunQueue;
|
|
96
|
+
export declare function writeRunQueue(path: string, queue: RunQueue): Digest;
|
|
97
|
+
export declare function readRunQueue(path: string, experiment?: ExperimentConfiguration, options?: ValidateRunQueueOptions): RunQueue;
|
|
98
|
+
export declare function readBoundedFile(path: string, label?: string, afterOpen?: () => void, maxBytes?: number): Buffer;
|
|
99
|
+
export declare function validateRunQueue(queue: unknown, experiment?: ExperimentConfiguration, artifact?: string, options?: ValidateRunQueueOptions): ArtifactValidationError[];
|
|
100
|
+
export declare function assertValidRunQueue(queue: unknown, experiment?: ExperimentConfiguration, artifact?: string, options?: ValidateRunQueueOptions): asserts queue is RunQueue;
|
|
101
|
+
export declare function inspectRunQueue(queue: RunQueue): RunQueueInspection;
|
|
102
|
+
export declare class LocalRunQueue {
|
|
103
|
+
private position;
|
|
104
|
+
private readonly snapshot;
|
|
105
|
+
constructor(queue: RunQueue);
|
|
106
|
+
get queue(): RunQueue;
|
|
107
|
+
get remaining(): number;
|
|
108
|
+
next(): RunQueueEntry | undefined;
|
|
109
|
+
}
|