Files
openclaw/src/agents/transcript-policy.ts
Peter Steinberger bb46b79d3c refactor: internalize OpenClaw agent runtime (#85341)
* refactor: extract agent core package

Introduce packages/agent-core as the OpenClaw-owned home for reusable agent loop, harness, session, prompt, and runtime dependency contracts.

* refactor: extract shared llm runtime

Move provider model registries, stream wrappers, OAuth helpers, and LLM utilities into src/llm with plugin-sdk barrels instead of depending on the old embedded runtime layout.

* refactor: remove pi runtime internals

Rename remaining Pi-shaped agent surfaces to OpenClaw agent runtime names, delete obsolete Pi docs and package graph checks, and add the third-party notice for incorporated code.

* refactor: tighten agent session runtime

Make agent-core/runtime dependencies explicit, consolidate compaction and session transcript helpers, and move model/session helpers behind OpenClaw-owned contracts.

* refactor: remove static model and pi auth paths

Drop static model catalogs and Pi auth bridges, move model/provider facts to manifest-owned runtime contracts, and harden internal embedded-agent utilities.

* refactor: remove legacy provider compat paths

* docs: remove agent parity notes

* fix: skip provider wildcard metadata parsing

* refactor: share session extension sdk loading

* refactor: inline acpx proxy error formatter

* refactor: fold edit recovery into edit tool

* fix: accept extension batch separator

* test: align startup provider plugin expectations

* fix: restore provider-scoped release discovery

* test: align static asset packaging expectations

* fix: run static provider catalogs during scoped discovery

* fix: add provider entry catalogs for scoped live discovery

* fix: load lightweight provider catalog entries

* fix: refresh provider-scoped plugin metadata

* fix: keep provider catalog entries on release live path

* fix: keep static manifest models in release live checks

* fix: harden release model discovery

* fix: reduce OpenAI live cache probe reasoning

* fix: disable OpenAI cache probe reasoning

* ci: extend OpenAI gateway live timeout

* fix: extend live gateway model budget

* fix: stabilize release validation regressions

* fix: honor provider aliases in model rows

* fix: stabilize release validation lanes

* fix: stabilize release memory qa

* ci: stabilize release validation lanes

* ci: prefer ipv4 for live docker node calls

* fix: restore shared tool-call stream wrapper

* ci: remove legacy pi test shard alias

* fix: clean up embedded agent test drift

* fix: stabilize runtime alias status

* fix: clean up embedded agent ci drift

* fix: restore release ci invariants

* fix: clean up post-rebase runtime drift

* fix: restore release ci checks

* fix: restore release ci after rebase

* fix: remove stale pi runtime path

* test: align compaction runtime expectations

* test: update plugin prerelease expectations

* fix: handle claude live tool approvals

* fix: stabilize release validation gates

* fix: finish agent runtime import

* test: finish post-rebase agent runtime mocks

* fix: keep codex compaction native

* fix: stabilize codex app-server hook tests

* test: isolate codex diagnostic active run

* test: remove codex diagnostic completion race

# Conflicts:
#	extensions/codex/src/app-server/run-attempt.test.ts

* ci: fix full release manifest performance run id

* refactor: narrow llm plugin sdk boundary

* chore: drop generated google boundary stamps

* fix: repair rebase fallout

* fix: clean up rebased runtime references

* fix: decode codex jwt payloads as base64url

* fix: preserve shipped pi runtime alias

* fix: add scoped sdk virtual modules

* fix: decode llm codex oauth jwt as base64url

* fix: avoid stale vertex adc negative cache

* fix: harden tool arg decoding and codeql path

* fix: keep vertex adc negative checks live

* refactor: consolidate codex jwt and edit helpers

* fix: await codex oauth node runtime imports

* fix: preserve sdk tool and notice contracts

* fix: preserve shipped compat config boundaries

* fix: align codex oauth callback host

* fix: terminate agent-core loop streams on failure

* fix: keep codex oauth callback alive during fallback

* ci: include session tools in critical codeql scans

* fix: keep Cloudflare Anthropic provider auth header

* docs: redirect legacy pi runtime pages

* fix: honor bundled web provider compat discovery

* fix: protect session output spill files

* fix: keep legacy agent dir env blocked

* fix: contain auto-discovered skill symlinks

* fix: harden agent core sdk proxy surfaces

* fix: restore approval reaction sdk compat

* fix: keep live docker runs bounded

* fix: keep codex oauth redirect host aligned

* fix: resolve post-rebase agent runtime drift

* fix: redact anthropic oauth parse failures

* fix: preserve responses strict tool shaping

* fix: repair agent runtime rebase cleanup

* docs: redirect retired parity pages

* fix: bound auto-discovered resources to roots

* fix: repair post-rebase agent test drift

* fix: preserve bundled provider allowlist migration

* fix: preserve manifest-owned provider aliases

* fix: declare photon image dependency

* fix: keep provider headers out of proxy body

* fix: preserve shipped env aliases

* fix: refresh control ui i18n generated state

* fix: quote read fallback paths

* fix: preview edits through configured backend

* test: satisfy core test typecheck

* fix: preserve ZAI usage auth fallback

* test: repair codex diagnostic test

* fix: repair agent runtime rebase drift

* test: finish embedded runner import rename

* fix: repair agent runtime rebase integrations

* test: align compaction oauth fallback expectations

* fix: allow sdk-auth session models

* fix: update doctor tool schema import

* fix: preserve bedrock plugin region

* fix: stream harmony-like prose immediately

* ci: include session runtime in codeql shards

* fix: repair latest rebase integrations

* fix: honor explicit codex websocket transport

* fix: keep openai-compatible credentials provider-scoped

* fix: refresh sdk api baseline after rebase

* fix: route cli runtime aliases through openclaw harness

* test: rename stale harness mock expectation

* test: rename embedded agent overflow calls

* test: clean embedded auth test wording

* test: use openclaw stream types in deepinfra cache test

* fix: refresh sdk api baseline on latest main

* fix: honor bundled discovery compat allowlists

* fix: refresh sdk api baseline after latest rebase

* fix: remove stale rebase imports

* test: rename stale model catalog mock

* test: mock renamed doctor runtime modules

* fix: map canonical kimi env auth

* fix: use internal model registry in bench script

* fix: migrate deepinfra provider catalog entry

* fix: enforce builtin tool suppression

* fix: route compaction auth and proxy payloads safely

* refactor: prune unused llm registry leftovers

* test: update codex hooks session import

* test: fix model picker ci coverage

* test: align model picker auth mock types
2026-05-27 19:24:04 +01:00

344 lines
12 KiB
TypeScript

import type { OpenClawConfig } from "../config/types.openclaw.js";
import { resolvePluginControlPlaneFingerprint } from "../plugins/plugin-control-plane-context.js";
import type { ProviderRuntimePluginHandle } from "../plugins/provider-hook-runtime.js";
import { resolveProviderRuntimePlugin } from "../plugins/provider-hook-runtime.js";
import { shouldPreserveThinkingBlocks } from "../plugins/provider-replay-helpers.js";
import type { ProviderRuntimeModel } from "../plugins/provider-runtime-model.types.js";
import type { ProviderReplayPolicy } from "../plugins/types.js";
import { normalizeLowercaseStringOrEmpty } from "../shared/string-coerce.js";
import { isGoogleModelApi } from "./embedded-agent-helpers/google.js";
import { normalizeProviderId } from "./model-selection.js";
import type { ToolCallIdMode } from "./tool-call-id.js";
export type TranscriptSanitizeMode = "full" | "images-only";
export type TranscriptPolicy = {
sanitizeMode: TranscriptSanitizeMode;
sanitizeToolCallIds: boolean;
toolCallIdMode?: ToolCallIdMode;
preserveNativeAnthropicToolUseIds: boolean;
repairToolUseResultPairing: boolean;
preserveSignatures: boolean;
sanitizeThoughtSignatures?: {
allowBase64Only?: boolean;
includeCamelCase?: boolean;
};
sanitizeThinkingSignatures: boolean;
dropThinkingBlocks: boolean;
dropReasoningFromHistory?: boolean;
applyGoogleTurnOrdering: boolean;
validateGeminiTurns: boolean;
validateAnthropicTurns: boolean;
allowSyntheticToolResults: boolean;
};
const SIGNED_THINKING_PROVIDERS = new Set(["anthropic", "amazon-bedrock", "anthropic-vertex"]);
export function providerRequiresSignedThinking(provider?: string | null): boolean {
return SIGNED_THINKING_PROVIDERS.has(normalizeProviderId(provider ?? ""));
}
export function shouldAllowProviderOwnedThinkingReplay(params: {
modelApi?: string | null;
provider?: string | null;
policy: Pick<
TranscriptPolicy,
"validateAnthropicTurns" | "preserveSignatures" | "dropThinkingBlocks"
>;
}): boolean {
const hasProviderOwnedSignedThinking =
params.policy.preserveSignatures || providerRequiresSignedThinking(params.provider);
return (
isAnthropicApi(params.modelApi) &&
params.policy.validateAnthropicTurns &&
hasProviderOwnedSignedThinking &&
!params.policy.dropThinkingBlocks
);
}
const DEFAULT_TRANSCRIPT_POLICY: TranscriptPolicy = {
sanitizeMode: "images-only",
sanitizeToolCallIds: false,
toolCallIdMode: undefined,
preserveNativeAnthropicToolUseIds: false,
repairToolUseResultPairing: true,
preserveSignatures: false,
sanitizeThoughtSignatures: undefined,
sanitizeThinkingSignatures: false,
dropThinkingBlocks: false,
dropReasoningFromHistory: false,
applyGoogleTurnOrdering: false,
validateGeminiTurns: false,
validateAnthropicTurns: false,
allowSyntheticToolResults: false,
};
function isAnthropicApi(modelApi?: string | null): boolean {
return modelApi === "anthropic-messages" || modelApi === "bedrock-converse-stream";
}
function isOpenAiResponsesCompatibleApi(modelApi?: string | null): boolean {
return (
modelApi === "openai-responses" ||
modelApi === "openai-codex-responses" ||
modelApi === "azure-openai-responses"
);
}
function isClaudeFamilyModelId(modelId?: string | null): boolean {
const id = normalizeLowercaseStringOrEmpty(modelId);
return /(?:^|[./:_-])claude(?:$|[./:_-])/.test(id);
}
function modelDisablesReasoningEffort(model?: ProviderRuntimeModel): boolean {
const compat = model?.compat as { supportsReasoningEffort?: boolean } | undefined;
return compat?.supportsReasoningEffort === false;
}
/**
* Provides a narrow replay-policy fallback for providers that do not have an
* owning runtime plugin.
*
* This exists to preserve generic custom-provider behavior. Bundled providers
* should express replay ownership through `buildReplayPolicy` instead.
*/
function buildUnownedProviderTransportReplayFallback(params: {
modelApi?: string | null;
modelId?: string | null;
model?: ProviderRuntimeModel;
}): ProviderReplayPolicy | undefined {
const isGoogle = isGoogleModelApi(params.modelApi);
const isAnthropic = isAnthropicApi(params.modelApi);
const isStrictOpenAiCompatible = params.modelApi === "openai-completions";
const requiresOpenAiCompatibleToolIdSanitization =
params.modelApi === "openai-completions" ||
params.modelApi === "openai-responses" ||
params.modelApi === "openai-codex-responses" ||
params.modelApi === "azure-openai-responses";
if (
!isGoogle &&
!isAnthropic &&
!isStrictOpenAiCompatible &&
!requiresOpenAiCompatibleToolIdSanitization
) {
return undefined;
}
const modelId = normalizeLowercaseStringOrEmpty(params.modelId);
const isClaudeOpenAiResponses = isOpenAiResponsesCompatibleApi(params.modelApi)
? isClaudeFamilyModelId(modelId)
: false;
return {
...(isGoogle || isAnthropic ? { sanitizeMode: "full" as const } : {}),
...(isGoogle || isAnthropic || requiresOpenAiCompatibleToolIdSanitization
? {
sanitizeToolCallIds: true,
toolCallIdMode: "strict" as const,
}
: {}),
...(isAnthropic ? { preserveSignatures: true } : {}),
...(isGoogle
? {
sanitizeThoughtSignatures: {
allowBase64Only: true,
includeCamelCase: true,
},
}
: {}),
...(isAnthropic && modelId.includes("claude")
? { dropThinkingBlocks: !shouldPreserveThinkingBlocks(modelId) }
: {}),
...(isAnthropic && modelDisablesReasoningEffort(params.model)
? { dropThinkingBlocks: true }
: {}),
...(isStrictOpenAiCompatible
? { dropReasoningFromHistory: !requiresReasoningContentReplay(params.modelId) }
: {}),
...(isGoogle || isStrictOpenAiCompatible ? { applyAssistantFirstOrderingFix: true } : {}),
...(isGoogle || isStrictOpenAiCompatible ? { validateGeminiTurns: true } : {}),
...(isAnthropic || isStrictOpenAiCompatible || isClaudeOpenAiResponses
? { validateAnthropicTurns: true }
: {}),
...(isGoogle || isAnthropic || isOpenAiResponsesCompatibleApi(params.modelApi)
? { allowSyntheticToolResults: true }
: {}),
};
}
const REASONING_CONTENT_REPLAY_MODEL_IDS = new Set([
"kimi-for-coding",
"kimi-k2.5",
"kimi-k2.6",
"kimi-k2-thinking",
"kimi-k2-thinking-turbo",
"mimo-v2-pro",
"mimo-v2-omni",
"mimo-v2.5",
"mimo-v2.5-pro",
"mimo-v2.6-pro",
]);
function requiresReasoningContentReplay(modelId: string | null | undefined): boolean {
const normalized = normalizeLowercaseStringOrEmpty(modelId);
if (!normalized) {
return false;
}
const parts = normalized.split("/").filter(Boolean);
const finalPart = parts[parts.length - 1] ?? normalized;
const candidates = [finalPart];
const colonParts = finalPart.split(":").filter(Boolean);
if (colonParts.length > 1) {
candidates.push(colonParts[0] ?? "", colonParts[colonParts.length - 1] ?? "");
}
return candidates.some((candidate) => REASONING_CONTENT_REPLAY_MODEL_IDS.has(candidate));
}
function mergeTranscriptPolicy(
policy: ProviderReplayPolicy | undefined,
basePolicy: TranscriptPolicy = DEFAULT_TRANSCRIPT_POLICY,
): TranscriptPolicy {
if (!policy) {
return basePolicy;
}
return {
...basePolicy,
...(policy.sanitizeMode != null ? { sanitizeMode: policy.sanitizeMode } : {}),
...(typeof policy.sanitizeToolCallIds === "boolean"
? { sanitizeToolCallIds: policy.sanitizeToolCallIds }
: {}),
...(policy.toolCallIdMode ? { toolCallIdMode: policy.toolCallIdMode as ToolCallIdMode } : {}),
...(typeof policy.preserveNativeAnthropicToolUseIds === "boolean"
? { preserveNativeAnthropicToolUseIds: policy.preserveNativeAnthropicToolUseIds }
: {}),
...(typeof policy.repairToolUseResultPairing === "boolean"
? { repairToolUseResultPairing: policy.repairToolUseResultPairing }
: {}),
...(typeof policy.preserveSignatures === "boolean"
? { preserveSignatures: policy.preserveSignatures }
: {}),
...(policy.sanitizeThoughtSignatures
? { sanitizeThoughtSignatures: policy.sanitizeThoughtSignatures }
: {}),
...(typeof policy.dropThinkingBlocks === "boolean"
? { dropThinkingBlocks: policy.dropThinkingBlocks }
: {}),
...(typeof policy.dropReasoningFromHistory === "boolean"
? { dropReasoningFromHistory: policy.dropReasoningFromHistory }
: {}),
...(typeof policy.applyAssistantFirstOrderingFix === "boolean"
? { applyGoogleTurnOrdering: policy.applyAssistantFirstOrderingFix }
: {}),
...(typeof policy.validateGeminiTurns === "boolean"
? { validateGeminiTurns: policy.validateGeminiTurns }
: {}),
...(typeof policy.validateAnthropicTurns === "boolean"
? { validateAnthropicTurns: policy.validateAnthropicTurns }
: {}),
...(typeof policy.allowSyntheticToolResults === "boolean"
? { allowSyntheticToolResults: policy.allowSyntheticToolResults }
: {}),
};
}
const transcriptPolicyCache = new WeakMap<OpenClawConfig, Map<string, TranscriptPolicy>>();
function canCacheTranscriptPolicy(params: {
config?: OpenClawConfig;
env?: NodeJS.ProcessEnv;
}): params is { config: OpenClawConfig; env?: NodeJS.ProcessEnv } {
if (!params.config) {
return false;
}
return !params.env || params.env === process.env;
}
function resolveTranscriptPolicyCacheKey(params: {
modelApi?: string | null;
provider: string;
modelId?: string | null;
model?: ProviderRuntimeModel;
config: OpenClawConfig;
workspaceDir?: string;
env?: NodeJS.ProcessEnv;
}): string {
return JSON.stringify({
provider: params.provider,
modelApi: params.modelApi ?? "",
modelId: params.modelId ?? "",
dropsThinkingForReasoningCompat: modelDisablesReasoningEffort(params.model),
workspaceDir: params.workspaceDir ?? "",
pluginControlPlane: resolvePluginControlPlaneFingerprint({
config: params.config,
workspaceDir: params.workspaceDir,
env: params.env,
}),
});
}
export function resolveTranscriptPolicy(params: {
modelApi?: string | null;
provider?: string | null;
modelId?: string | null;
config?: OpenClawConfig;
workspaceDir?: string;
env?: NodeJS.ProcessEnv;
model?: ProviderRuntimeModel;
runtimeHandle?: ProviderRuntimePluginHandle;
}): TranscriptPolicy {
const provider = normalizeProviderId(params.provider ?? "");
const cacheConfig = canCacheTranscriptPolicy(params) ? params.config : undefined;
const cacheKey = cacheConfig
? resolveTranscriptPolicyCacheKey({ ...params, provider, config: cacheConfig })
: undefined;
if (cacheConfig && cacheKey) {
const cached = transcriptPolicyCache.get(cacheConfig)?.get(cacheKey);
if (cached) {
return cached;
}
}
const runtimePlugin =
params.runtimeHandle?.plugin ??
(provider
? resolveProviderRuntimePlugin({
provider,
modelId: params.modelId,
config: params.config,
workspaceDir: params.workspaceDir,
env: params.env,
})
: undefined);
const context = {
config: params.config,
workspaceDir: params.workspaceDir,
env: params.env,
provider,
modelId: params.modelId ?? "",
modelApi: params.modelApi,
model: params.model,
};
// Once a provider adopts the replay-policy hook, replay policy should come
// from the plugin, not from transport-family defaults in core.
const buildReplayPolicy = runtimePlugin?.buildReplayPolicy;
const policy = buildReplayPolicy
? mergeTranscriptPolicy(buildReplayPolicy(context) ?? undefined)
: mergeTranscriptPolicy(
buildUnownedProviderTransportReplayFallback({
modelApi: params.modelApi,
modelId: params.modelId,
model: params.model,
}),
);
if (cacheConfig && cacheKey) {
let configCache = transcriptPolicyCache.get(cacheConfig);
if (!configCache) {
configCache = new Map();
transcriptPolicyCache.set(cacheConfig, configCache);
}
configCache.set(cacheKey, policy);
}
return policy;
}