Never stop coding. Free MIT AI gateway: one endpoint, 375 providers (150+ free), 1,400+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/PWA. Built by 900+ contributors
TypeScript a2aai-agentsai-gatewayanthropicclaudeclaude-codeclinecodexcopilotcursordeepseekfree-aigeminikimillm-gatewaymcpopenaiopenai-proxyqwentoken-saver
import { CREDENTIAL_PATTERNS } from "@omniroute/open-sse/utils/credentialPatterns.ts";
import { getSettings } from "@/lib/db/settings";
import { BaseGuardrail, type GuardrailContext, type GuardrailResult } from "./base";
export { CREDENTIAL_PATTERNS };
export type { CredentialPattern } from "@omniroute/open-sse/utils/credentialPatterns.ts";
/**
* CredentialMaskerGuardrail — redacts well-known API-key / secret-token patterns
* from the upstream payload (message content, tool-call arguments, tool results)
* AND the provider response, so secrets are not leaked to providers or clients.
*
* Opt-in: enabled when CREDENTIAL_REDACTION_ENABLED=true (mirrors PII_REDACTION_ENABLED).
* Patterns are provider-specific + conservative to avoid false positives.
* Future: per-pipeline / per-provider scoping via GuardrailContext.
*/
export interface CredentialRedactionResult {
text: string;
detections: Array<{ type: string; count: number }>;
modified: boolean;
}
export function redactCredentials(text: string): CredentialRedactionResult {
if (typeof text !== "string" || !text) return { text, detections: [], modified: false };
let result = text;
const detections: Array<{ type: string; count: number }> = [];
for (const p of CREDENTIAL_PATTERNS) {
p.regex.lastIndex = 0;
const matches = result.match(p.regex);
if (matches && matches.length > 0) {
result = result.replace(p.regex, p.replacement);
detections.push({ type: p.name, count: matches.length });
}
}
return { text: result, detections, modified: result !== text };
}
type JsonRecord = Record<string, unknown>;
function isSensitiveHeaderKey(key: string): boolean {
return ["authorization", "x-api-key", "api-key", "apikey"].includes(key.toLowerCase());
}
function redactHeaderValue(value: string): string {
const schemePrefix = value.match(/^(\s*(?:(?:Bearer|Basic|Token)\s+)?)/i)?.[1] || "";
return schemePrefix + "[REDACTED:auth_header]";
}
/** Redact values without cloning unchanged branches or non-plain objects. */
function walkValue(
value: unknown,
detections: Array<{ type: string; count: number }>,
seen = new WeakSet<object>()
): { modified: boolean; value: unknown } {
if (typeof value === "string") {
// #13462: Skip redaction for base64-encoded image data URLs — applying
// credential regexes to binary transport data corrupts valid base64.
if (isImageDataUrl(value)) return { modified: false, value };
const r = redactCredentials(value);
if (r.detections.length) detections.push(...r.detections);
return { modified: r.modified, value: r.text };
}
if (Array.isArray(value)) {
if (seen.has(value)) return { modified: false, value };
seen.add(value);
let next: unknown[] | null = null;
for (let index = 0; index < value.length; index++) {
const r = walkValue(value[index], detections, seen);
if (r.modified && !next) next = value.slice(0, index);
if (next) next[index] = r.value;
}
seen.delete(value);
return { modified: next !== null, value: next ?? value };
}
if (value && typeof value === "object") {
if (seen.has(value)) return { modified: false, value };
const prototype = Object.getPrototypeOf(value);
if (prototype !== null && prototype !== Object.prototype) return { modified: false, value };
seen.add(value);
const entries = Object.entries(value as JsonRecord);
let next: JsonRecord | null = null;
for (let index = 0; index < entries.length; index++) {
const [key, entryValue] = entries[index];
const structuredHeader = isSensitiveHeaderKey(key) && typeof entryValue === "string";
const redactedHeader = structuredHeader ? redactHeaderValue(entryValue) : null;
const r = structuredHeader
? { modified: redactedHeader !== entryValue, value: redactedHeader }
: walkValue(entryValue, detections, seen);
if (structuredHeader && r.modified) detections.push({ type: "auth_header", count: 1 });
if (r.modified && !next) {
next = Object.create(prototype) as JsonRecord;
for (let previous = 0; previous < index; previous++) {
const [previousKey, previousValue] = entries[previous];
next[previousKey] = previousValue;
}
}
if (next) next[key] = r.value;
}
seen.delete(value);
return { modified: next !== null, value: next ?? value };
}
return { modified: false, value };
}
/**
* Check if a string is a base64-encoded image data URL that should not be
* subject to text-pattern redaction. Applying credential regexes to binary
* transport data causes false-positive corruption of valid base64 content.
*/
function isImageDataUrl(value: string): boolean {
// Match data URLs with image MIME types (data:image/png;base64,... etc.)
return /^data:image\/[a-zA-Z0-9.+-]+;base64,/.test(value);
}
/** Walk request payloads without changing safe values. */
function redactPayload(
payload: unknown,
detections: Array<{ type: string; count: number }>
): { modified: boolean; payload: unknown } {
const r = walkValue(payload, detections);
return { modified: r.modified, payload: r.value };
}
/** Walk provider responses and redact credentials. */
function redactResponse(
response: unknown,
detections: Array<{ type: string; count: number }>
): { modified: boolean; response: unknown } {
const r = walkValue(response, detections);
return { modified: r.modified, response: r.value };
}
export class CredentialMaskerGuardrail extends BaseGuardrail {
constructor(options: { enabled?: boolean; priority?: number } = {}) {
super("credential-masker", { enabled: options.enabled, priority: options.priority ?? 95 });
}
async preCall(
payload: unknown,
_context: GuardrailContext
): Promise<GuardrailResult<unknown> | void> {
const _s = await getSettings();
if (!(
_s.credentialRedactionEnabled === true || process.env.CREDENTIAL_REDACTION_ENABLED === "true"
))
return { block: false };
const detections: Array<{ type: string; count: number }> = [];
const { modified, payload: next } = redactPayload(payload, detections);
if (!modified) return { block: false };
return {
block: false,
modifiedPayload: next,
meta: { credentialsRedacted: detections, count: detections.reduce((n, d) => n + d.count, 0) },
};
}
async postCall(
response: unknown,
_context: GuardrailContext
): Promise<GuardrailResult<unknown> | void> {
const _s = await getSettings();
if (!(
_s.credentialRedactionEnabled === true || process.env.CREDENTIAL_REDACTION_ENABLED === "true"
))
return { block: false };
const detections: Array<{ type: string; count: number }> = [];
const { modified, response: next } = redactResponse(response, detections);
if (!modified) return { block: false };
return {
block: false,
modifiedResponse: next,
meta: { credentialsRedacted: detections, count: detections.reduce((n, d) => n + d.count, 0) },
};
}
}