revert(tools): undo accidental merge of PR #18584

This commit is contained in:
Sebastian
2026-02-16 21:13:03 -05:00
parent 0158e41298
commit f924ab40d8
7 changed files with 8 additions and 107 deletions

View File

@@ -1,16 +1,12 @@
import { Type } from "@sinclair/typebox";
import type { OpenClawConfig } from "../../config/config.js";
import type { AnyAgentTool } from "./common.js";
import { fetchWithSsrFGuard } from "../../infra/net/fetch-guard.js";
import {
matchesHostnameAllowlist,
normalizeHostnameAllowlist,
SsrFBlockedError,
} from "../../infra/net/ssrf.js";
import { SsrFBlockedError } from "../../infra/net/ssrf.js";
import { logDebug } from "../../logger.js";
import { wrapExternalContent, wrapWebContent } from "../../security/external-content.js";
import { normalizeSecretInput } from "../../utils/normalize-secret-input.js";
import { stringEnum } from "../schema/typebox.js";
import type { AnyAgentTool } from "./common.js";
import { jsonResult, readNumberParam, readStringParam } from "./common.js";
import {
extractReadableContent,
@@ -26,7 +22,6 @@ import {
normalizeCacheKey,
readCache,
readResponseText,
resolveWebUrlAllowlist,
resolveCacheTtlMs,
resolveTimeoutSeconds,
withTimeout,
@@ -73,22 +68,6 @@ type WebFetchConfig = NonNullable<OpenClawConfig["tools"]>["web"] extends infer
: undefined
: undefined;
type WebConfig = NonNullable<OpenClawConfig["tools"]>["web"];
export function resolveFetchUrlAllowlist(web?: WebConfig): string[] | undefined {
return resolveWebUrlAllowlist(web);
}
export function isUrlAllowedByAllowlist(url: string, allowlist: string[]): boolean {
try {
const hostname = new URL(url).hostname;
const normalizedAllowlist = normalizeHostnameAllowlist(allowlist);
return matchesHostnameAllowlist(hostname, normalizedAllowlist);
} catch {
return false;
}
}
type FirecrawlFetchConfig =
| {
enabled?: boolean;
@@ -753,7 +732,6 @@ export function createWebFetchTool(options?: {
(fetch && "userAgent" in fetch && typeof fetch.userAgent === "string" && fetch.userAgent) ||
DEFAULT_FETCH_USER_AGENT;
const maxResponseBytes = resolveFetchMaxResponseBytes(fetch);
const urlAllowlist = resolveFetchUrlAllowlist(options?.config?.tools?.web);
return {
label: "Web Fetch",
name: "web_fetch",
@@ -763,25 +741,6 @@ export function createWebFetchTool(options?: {
execute: async (_toolCallId, args) => {
const params = args as Record<string, unknown>;
const url = readStringParam(params, "url", { required: true });
// Check URL against allowlist if configured
if (urlAllowlist && urlAllowlist.length > 0) {
if (!isUrlAllowedByAllowlist(url, urlAllowlist)) {
let hostname: string;
try {
hostname = new URL(url).hostname;
} catch {
hostname = url;
}
return jsonResult({
error: "url_not_allowed",
message: `URL not in allowlist. Allowed domains: ${urlAllowlist.join(", ")}`,
blockedUrl: url,
blockedHostname: hostname,
});
}
}
const extractMode = readStringParam(params, "extractMode") === "text" ? "text" : "markdown";
const maxChars = readNumberParam(params, "maxChars", { integer: true });
const maxCharsCap = resolveFetchMaxCharsCap(fetch);

View File

@@ -1,10 +1,9 @@
import { Type } from "@sinclair/typebox";
import { formatCliCommand } from "../../cli/command-format.js";
import type { OpenClawConfig } from "../../config/config.js";
import { matchesHostnameAllowlist, normalizeHostnameAllowlist } from "../../infra/net/ssrf.js";
import type { AnyAgentTool } from "./common.js";
import { formatCliCommand } from "../../cli/command-format.js";
import { wrapWebContent } from "../../security/external-content.js";
import { normalizeSecretInput } from "../../utils/normalize-secret-input.js";
import type { AnyAgentTool } from "./common.js";
import { jsonResult, readNumberParam, readStringParam } from "./common.js";
import {
CacheEntry,
@@ -13,7 +12,6 @@ import {
normalizeCacheKey,
readCache,
readResponseText,
resolveWebUrlAllowlist,
resolveCacheTtlMs,
resolveTimeoutSeconds,
withTimeout,
@@ -77,33 +75,6 @@ type WebSearchConfig = NonNullable<OpenClawConfig["tools"]>["web"] extends infer
: undefined
: undefined;
type WebConfig = NonNullable<OpenClawConfig["tools"]>["web"];
export function resolveUrlAllowlist(web?: WebConfig): string[] | undefined {
return resolveWebUrlAllowlist(web);
}
export function filterResultsByAllowlist(
results: Array<{ url?: string; siteName?: string }>,
allowlist: string[],
): Array<{ url?: string; siteName?: string }> {
if (allowlist.length === 0) {
return results;
}
const normalizedAllowlist = normalizeHostnameAllowlist(allowlist);
return results.filter((result) => {
if (!result.url) {
return true; // Keep entries without URL
}
try {
const hostname = new URL(result.url).hostname;
return matchesHostnameAllowlist(hostname, normalizedAllowlist);
} catch {
return true; // Keep entries with invalid URLs (let them pass through)
}
});
}
type BraveSearchResult = {
title?: string;
url?: string;
@@ -595,7 +566,6 @@ async function runWebSearch(params: {
perplexityModel?: string;
grokModel?: string;
grokInlineCitations?: boolean;
urlAllowlist?: string[];
}): Promise<Record<string, unknown>> {
const cacheKey = normalizeCacheKey(
params.provider === "brave"
@@ -718,15 +688,10 @@ async function runWebSearch(params: {
};
});
// Filter results by urlAllowlist if configured
const filteredResults = params.urlAllowlist
? filterResultsByAllowlist(mapped, params.urlAllowlist)
: mapped;
const payload = {
query: params.query,
provider: params.provider,
count: filteredResults.length,
count: mapped.length,
tookMs: Date.now() - start,
externalContent: {
untrusted: true,
@@ -734,7 +699,7 @@ async function runWebSearch(params: {
provider: params.provider,
wrapped: true,
},
results: filteredResults,
results: mapped,
};
writeCache(SEARCH_CACHE, cacheKey, payload, params.cacheTtlMs);
return payload;
@@ -752,7 +717,6 @@ export function createWebSearchTool(options?: {
const provider = resolveSearchProvider(search);
const perplexityConfig = resolvePerplexityConfig(search);
const grokConfig = resolveGrokConfig(search);
const urlAllowlist = resolveUrlAllowlist(options?.config?.tools?.web);
const description =
provider === "perplexity"
@@ -822,7 +786,6 @@ export function createWebSearchTool(options?: {
perplexityModel: resolvePerplexityModel(perplexityConfig),
grokModel: resolveGrokModel(grokConfig),
grokInlineCitations: resolveGrokInlineCitations(grokConfig),
urlAllowlist,
});
return jsonResult(result);
},
@@ -840,6 +803,4 @@ export const __testing = {
resolveGrokModel,
resolveGrokInlineCitations,
extractGrokContent,
resolveUrlAllowlist,
filterResultsByAllowlist,
} as const;

View File

@@ -8,20 +8,6 @@ export const DEFAULT_TIMEOUT_SECONDS = 30;
export const DEFAULT_CACHE_TTL_MINUTES = 15;
const DEFAULT_CACHE_MAX_ENTRIES = 100;
export function resolveWebUrlAllowlist(web: unknown): string[] | undefined {
if (!web || typeof web !== "object") {
return undefined;
}
if (!("urlAllowlist" in web)) {
return undefined;
}
const allowlist = (web as { urlAllowlist?: unknown }).urlAllowlist;
if (!Array.isArray(allowlist)) {
return undefined;
}
return allowlist.length > 0 ? allowlist : undefined;
}
export function resolveTimeoutSeconds(value: unknown, fallback: number): number {
const parsed = typeof value === "number" && Number.isFinite(value) ? value : fallback;
return Math.max(1, Math.floor(parsed));

View File

@@ -116,8 +116,6 @@ export const FIELD_HELP: Record<string, string> = {
"Perplexity base URL override (default: https://openrouter.ai/api/v1 or https://api.perplexity.ai).",
"tools.web.search.perplexity.model":
'Perplexity model override (default: "perplexity/sonar-pro").',
"tools.web.urlAllowlist":
"Optional URL/domain allowlist shared by web_search and web_fetch. Accepts domain patterns like 'example.com', '*.github.com'. When configured, only matching URLs are allowed.",
"tools.web.fetch.enabled": "Enable the web_fetch tool (lightweight HTTP fetch).",
"tools.web.fetch.maxChars": "Max characters returned by web_fetch (truncated).",
"tools.web.fetch.maxCharsCap":

View File

@@ -395,8 +395,6 @@ export type ToolsConfig = {
/** Optional tool policy overrides keyed by provider id or "provider/model". */
byProvider?: Record<string, ToolPolicyConfig>;
web?: {
/** Optional URL/domain allowlist for web tools. When configured, only URLs matching these patterns are allowed. */
urlAllowlist?: string[];
search?: {
/** Enable web search tool (default: true when API key is present). */
enabled?: boolean;

View File

@@ -267,7 +267,6 @@ export const ToolsWebFetchSchema = z
export const ToolsWebSchema = z
.object({
urlAllowlist: z.array(z.string()).optional(),
search: ToolsWebSearchSchema,
fetch: ToolsWebFetchSchema,
})

View File

@@ -33,7 +33,7 @@ function normalizeHostnameSet(values?: string[]): Set<string> {
return new Set(values.map((value) => normalizeHostname(value)).filter(Boolean));
}
export function normalizeHostnameAllowlist(values?: string[]): string[] {
function normalizeHostnameAllowlist(values?: string[]): string[] {
if (!values || values.length === 0) {
return [];
}
@@ -57,7 +57,7 @@ function isHostnameAllowedByPattern(hostname: string, pattern: string): boolean
return hostname === pattern;
}
export function matchesHostnameAllowlist(hostname: string, allowlist: string[]): boolean {
function matchesHostnameAllowlist(hostname: string, allowlist: string[]): boolean {
if (allowlist.length === 0) {
return true;
}