fix(agents): make image sanitization dimension configurable

This commit is contained in:
Peter Steinberger
2026-02-18 00:43:31 +01:00
parent 5ee79f80eb
commit b05e89e5e6
21 changed files with 156 additions and 45 deletions

View File

@@ -1,5 +1,3 @@
import fs from "node:fs/promises";
import os from "node:os";
import type { AgentMessage } from "@mariozechner/pi-agent-core";
import {
createAgentSession,
@@ -7,10 +5,14 @@ import {
SessionManager,
SettingsManager,
} from "@mariozechner/pi-coding-agent";
import { resolveHeartbeatPrompt } from "../../auto-reply/heartbeat.js";
import fs from "node:fs/promises";
import os from "node:os";
import type { ReasoningLevel, ThinkLevel } from "../../auto-reply/thinking.js";
import { resolveChannelCapabilities } from "../../config/channel-capabilities.js";
import type { OpenClawConfig } from "../../config/config.js";
import type { ExecElevatedDefaults } from "../bash-tools.js";
import type { EmbeddedPiCompactResult } from "./types.js";
import { resolveHeartbeatPrompt } from "../../auto-reply/heartbeat.js";
import { resolveChannelCapabilities } from "../../config/channel-capabilities.js";
import { getMachineDisplayName } from "../../infra/machine-name.js";
import { getGlobalHookRunner } from "../../plugins/hook-runner-global.js";
import { type enqueueCommand, enqueueCommandInLane } from "../../process/command-queue.js";
@@ -24,7 +26,6 @@ import { normalizeMessageChannel } from "../../utils/message-channel.js";
import { isReasoningTagProvider } from "../../utils/provider-utils.js";
import { resolveOpenClawAgentDir } from "../agent-paths.js";
import { resolveSessionAgentIds } from "../agent-scope.js";
import type { ExecElevatedDefaults } from "../bash-tools.js";
import { makeBootstrapWarn, resolveBootstrapContextForRun } from "../bootstrap-files.js";
import { listChannelSupportedActions, resolveChannelMessageToolHints } from "../channel-tools.js";
import { formatUserTime, resolveUserTimeFormat, resolveUserTimezone } from "../date-time.js";
@@ -81,7 +82,6 @@ import {
createSystemPromptOverride,
} from "./system-prompt.js";
import { splitSdkTools } from "./tool-split.js";
import type { EmbeddedPiCompactResult } from "./types.js";
import { describeUnknownError, mapThinkingLevel } from "./utils.js";
import { flushPendingToolResultsAfterIdle } from "./wait-for-idle-before-flush.js";
@@ -570,6 +570,7 @@ export async function compactEmbeddedPiSessionDirect(
modelApi: model.api,
modelId,
provider,
config: params.config,
sessionManager,
sessionId: params.sessionId,
policy: transcriptPolicy,

View File

@@ -1,12 +1,15 @@
import { EventEmitter } from "node:events";
import type { AgentMessage, AgentTool } from "@mariozechner/pi-agent-core";
import type { SessionManager } from "@mariozechner/pi-coding-agent";
import type { TSchema } from "@sinclair/typebox";
import { EventEmitter } from "node:events";
import type { OpenClawConfig } from "../../config/config.js";
import type { TranscriptPolicy } from "../transcript-policy.js";
import { registerUnhandledRejectionHandler } from "../../infra/unhandled-rejections.js";
import {
hasInterSessionUserProvenance,
normalizeInputProvenance,
} from "../../sessions/input-provenance.js";
import { resolveImageSanitizationLimits } from "../image-sanitization.js";
import {
downgradeOpenAIReasoningBlocks,
isCompactionFailureError,
@@ -20,7 +23,6 @@ import {
stripToolResultDetails,
sanitizeToolUseResultPairing,
} from "../session-transcript-repair.js";
import type { TranscriptPolicy } from "../transcript-policy.js";
import { resolveTranscriptPolicy } from "../transcript-policy.js";
import { log } from "./logger.js";
import { describeUnknownError } from "./utils.js";
@@ -416,6 +418,7 @@ export async function sanitizeSessionHistory(params: {
modelApi?: string | null;
modelId?: string;
provider?: string;
config?: OpenClawConfig;
sessionManager: SessionManager;
sessionId: string;
policy?: TranscriptPolicy;
@@ -438,6 +441,7 @@ export async function sanitizeSessionHistory(params: {
toolCallIdMode: policy.toolCallIdMode,
preserveSignatures: policy.preserveSignatures,
sanitizeThoughtSignatures: policy.sanitizeThoughtSignatures,
...resolveImageSanitizationLimits(params.config),
},
);
const sanitizedThinking = policy.normalizeAntigravityThinkingBlocks

View File

@@ -1,9 +1,10 @@
import fs from "node:fs/promises";
import os from "node:os";
import type { AgentMessage } from "@mariozechner/pi-agent-core";
import type { ImageContent } from "@mariozechner/pi-ai";
import { streamSimple } from "@mariozechner/pi-ai";
import { createAgentSession, SessionManager, SettingsManager } from "@mariozechner/pi-coding-agent";
import fs from "node:fs/promises";
import os from "node:os";
import type { EmbeddedRunAttemptParams, EmbeddedRunAttemptResult } from "./types.js";
import { resolveHeartbeatPrompt } from "../../../auto-reply/heartbeat.js";
import { resolveChannelCapabilities } from "../../../config/channel-capabilities.js";
import { getMachineDisplayName } from "../../../infra/machine-name.js";
@@ -33,6 +34,7 @@ import {
import { DEFAULT_CONTEXT_TOKENS } from "../../defaults.js";
import { resolveOpenClawDocsPath } from "../../docs-path.js";
import { isTimeoutError } from "../../failover-error.js";
import { resolveImageSanitizationLimits } from "../../image-sanitization.js";
import { resolveModelAuthMode } from "../../model-auth.js";
import { resolveDefaultModelForAgent } from "../../model-selection.js";
import { createOllamaStreamFn, OLLAMA_NATIVE_BASE_URL } from "../../ollama-stream.js";
@@ -105,7 +107,6 @@ import {
shouldFlagCompactionTimeout,
} from "./compaction-timeout.js";
import { detectAndLoadPromptImages } from "./images.js";
import type { EmbeddedRunAttemptParams, EmbeddedRunAttemptResult } from "./types.js";
export function injectHistoryImagesIntoMessages(
messages: AgentMessage[],
@@ -666,6 +667,7 @@ export async function runEmbeddedAttempt(
modelApi: params.model.api,
modelId: params.modelId,
provider: params.provider,
config: params.config,
sessionManager,
sessionId: params.sessionId,
policy: transcriptPolicy,
@@ -968,6 +970,7 @@ export async function runEmbeddedAttempt(
existingImages: params.images,
historyMessages: activeSession.messages,
maxBytes: MAX_IMAGE_BYTES,
maxDimensionPx: resolveImageSanitizationLimits(params.config).maxDimensionPx,
// Enforce sandbox path restrictions when sandbox is enabled
sandbox:
sandbox?.enabled && sandbox?.fsBridge

View File

@@ -1,9 +1,10 @@
import type { ImageContent } from "@mariozechner/pi-ai";
import path from "node:path";
import { fileURLToPath } from "node:url";
import type { ImageContent } from "@mariozechner/pi-ai";
import type { ImageSanitizationLimits } from "../../image-sanitization.js";
import type { SandboxFsBridge } from "../../sandbox/fs-bridge.js";
import { resolveUserPath } from "../../../utils.js";
import { loadWebMedia } from "../../../web/media.js";
import type { SandboxFsBridge } from "../../sandbox/fs-bridge.js";
import { sanitizeImageBlocks } from "../../tool-images.js";
import { log } from "../logger.js";
@@ -48,8 +49,13 @@ function isImageExtension(filePath: string): boolean {
async function sanitizeImagesWithLog(
images: ImageContent[],
label: string,
imageSanitization?: ImageSanitizationLimits,
): Promise<ImageContent[]> {
const { images: sanitized, dropped } = await sanitizeImageBlocks(images, label);
const { images: sanitized, dropped } = await sanitizeImageBlocks(
images,
label,
imageSanitization,
);
if (dropped > 0) {
log.warn(`Native image: dropped ${dropped} image(s) after sanitization (${label}).`);
}
@@ -354,6 +360,7 @@ export async function detectAndLoadPromptImages(params: {
existingImages?: ImageContent[];
historyMessages?: unknown[];
maxBytes?: number;
maxDimensionPx?: number;
sandbox?: { root: string; bridge: SandboxFsBridge };
}): Promise<{
/** Images for the current prompt (existingImages + detected in current prompt) */
@@ -437,10 +444,21 @@ export async function detectAndLoadPromptImages(params: {
}
}
const sanitizedPromptImages = await sanitizeImagesWithLog(promptImages, "prompt:images");
const imageSanitization: ImageSanitizationLimits = {
maxDimensionPx: params.maxDimensionPx,
};
const sanitizedPromptImages = await sanitizeImagesWithLog(
promptImages,
"prompt:images",
imageSanitization,
);
const sanitizedHistoryImagesByIndex = new Map<number, ImageContent[]>();
for (const [index, images] of historyImagesByIndex) {
const sanitized = await sanitizeImagesWithLog(images, `history:images:${index}`);
const sanitized = await sanitizeImagesWithLog(
images,
`history:images:${index}`,
imageSanitization,
);
if (sanitized.length > 0) {
sanitizedHistoryImagesByIndex.set(index, sanitized);
}