pi-freeflow 1.5.0 → 1.7.0

package/src/logger.ts

@@ -21,6 +21,18 @@ export const LOG_LEVEL_ORDER: Record<LogLevel, number> = {
 	audit: 4,
 };
 
+/**
+ * Levels that actually have emitters. 'audit' exists in LOG_LEVEL_ORDER as a
+ * sink for logAudit() callers but nothing decides the threshold by it — a
+ * persisted or env-provided 'audit' level would silently suppress everything
+ * below it (the audit-level freeze). Treat it as unknown everywhere.
+ */
+const EMITTED_LEVELS: readonly LogLevel[] = ["debug", "info", "warn", "error"];
+
+export function isEmittedLogLevel(level: string): level is LogLevel {
+	return (EMITTED_LEVELS as readonly string[]).includes(level);
+}
+
 let cachedDebugState: DebugState | null | undefined = undefined;
 let cachedDebugMtime = 0;
 let cachedDebugAt = 0;
@@ -50,10 +62,16 @@ export function loadDebugState(): DebugState | null {
 		const raw = fs.readFileSync(DEBUG_STATE_FILE, "utf8");
 		const parsed = JSON.parse(raw) as DebugState;
 		if (typeof parsed?.debug === "boolean") {
-			cachedDebugState = parsed;
+			// A stray 'audit' level (or any unknown level) must not suppress the
+			// whole log — keep the debug boolean, drop the unusable level.
+			const clean: DebugState = { debug: parsed.debug };
+			if (typeof parsed.level === "string" && isEmittedLogLevel(parsed.level)) {
+				clean.level = parsed.level;
+			}
+			cachedDebugState = clean;
 			cachedDebugMtime = stat.mtimeMs;
 			cachedDebugAt = now;
-			return parsed;
+			return clean;
 		}
 	} catch {
 		// Non-fatal if parsing or reading fails
@@ -91,6 +109,13 @@ export function saveDebugState(s: DebugState): void {
 	}
 }
 
+/** Test-only: drop the in-memory debug-state cache so a test's disk write is re-read. */
+export function _resetDebugStateCacheForTest(): void {
+	cachedDebugState = undefined;
+	cachedDebugMtime = 0;
+	cachedDebugAt = 0;
+}
+
 /**
  * Calculate the active minimum log level threshold based on state and env.
  */
@@ -99,13 +124,13 @@ export function getMinLogLevel(): number {
 	if (dbg?.debug) {
 		return LOG_LEVEL_ORDER.debug;
 	}
-	if (dbg?.level && dbg.level in LOG_LEVEL_ORDER) {
+	if (dbg?.level && isEmittedLogLevel(dbg.level)) {
 		return LOG_LEVEL_ORDER[dbg.level];
 	}
 
 	const raw = (process.env.FREEFLOW_LOG_LEVEL || "info").toLowerCase();
 
-	if (raw in LOG_LEVEL_ORDER) {
+	if (isEmittedLogLevel(raw)) {
 		return LOG_LEVEL_ORDER[raw as LogLevel];
 	}
 
@@ -143,9 +168,7 @@ export function rotateLogsIfNeeded(targetFile: string = LOG_FILE): void {
 			const dst = `${targetFile}.${i}`;
 			try {
 				if (fs.existsSync(src)) {
-					if (i === LOG_MAX_FILES && fs.existsSync(dst)) {
-						fs.unlinkSync(dst);
-					} else if (fs.existsSync(dst)) {
+					if (fs.existsSync(dst)) {
 						fs.unlinkSync(dst);
 					}
 					fs.renameSync(src, dst);
@@ -252,14 +275,6 @@ export function logError(
 	log("error", message, meta, reqId);
 }
 
-export function logAudit(
-	message: string,
-	meta?: Record<string, unknown>,
-	reqId?: string,
-): void {
-	log("audit", message, meta, reqId);
-}
-
 export interface ReadRecentLogsResult {
 	lines: string[];
 	totalMatched: number;
@@ -269,20 +284,28 @@ export interface ReadRecentLogsResult {
 
 /**
  * Read recent log entries from log file and its rotated archives.
+ * @param filterLevel Optional log-level filter (case-insensitive label match).
+ * @param filterReqId Optional request-ID substring filter.
+ * @param count Max lines to return (clamped 1-200).
+ * @param filterText Optional case-insensitive text substring filter.
+ * @param files Optional log files to read (newest-last order); defaults to
+ *   LOG_FILE and its rotated archives (.1-<LOG_MAX_FILES>).
  */
 export function readRecentLogs(
 	filterLevel?: LogLevel | null,
 	filterReqId?: string | null,
 	count = 25,
+	filterText?: string | null,
+	files?: string[],
 ): ReadRecentLogsResult {
-	const files: string[] = [
-		LOG_FILE,
-		`${LOG_FILE}.1`,
-		`${LOG_FILE}.2`,
-		`${LOG_FILE}.3`,
-	].filter((f) => fs.existsSync(f));
-
-	if (files.length === 0) {
+	const filesList: string[] = (
+		files ?? [
+			LOG_FILE,
+			...Array.from({ length: LOG_MAX_FILES }, (_, i) => `${LOG_FILE}.${i + 1}`),
+		]
+	).filter((f) => fs.existsSync(f));
+
+	if (filesList.length === 0) {
 		return {
 			lines: [],
 			totalMatched: 0,
@@ -292,7 +315,7 @@ export function readRecentLogs(
 	}
 
 	let allLines: string[] = [];
-	for (const f of files) {
+	for (const f of filesList) {
 		try {
 			const content = fs.readFileSync(f, "utf8");
 			const lines = content.trim().split("\n").filter(Boolean);
@@ -315,6 +338,11 @@ export function readRecentLogs(
 		);
 	}
 
+	if (filterText) {
+		const lower = filterText.toLowerCase();
+		filtered = filtered.filter((l) => l.toLowerCase().includes(lower));
+	}
+
 	const clampedCount = Math.min(200, Math.max(1, count));
 	const resultLines = filtered.slice(-clampedCount);
 

package/src/models.ts

@@ -1,12 +1,12 @@
 /**
  * Static model definitions and upstream routing catalogs for pi-freeflow
  *
- * Defines the 21 verified free models:
+ * Defines the 25 verified free models:
  * - 7 OpenCode Zen models (1 Responses API + 6 Chat Completions)
- * - 14 KiloCode Keyless Gateway models (10 OpenRouter format + 4 Standard format)
+ * - 18 KiloCode Keyless Gateway models (17 OpenRouter format + 1 Standard format)
  */
 
-import type { ModelDef, Upstream } from "./types.ts";
+import type { ModelDef, ThinkingLevelMap, Upstream } from "./types.ts";
 
 /**
  * OpenCode Zen free models verified against the live catalog and inference APIs.
@@ -21,13 +21,13 @@ export const OPENCODE_MODELS: ModelDef[] = [
 		maxTokens: 131_072,
 		api: "openai-responses",
 		input: ["text", "image"],
-		thinkingLevelMap: {
-			off: null,
-			minimal: "minimal",
-			low: "low",
-			medium: "medium",
-			high: "high",
-			xhigh: "xhigh",
+		thinkingLevelMap: {
+			off: null,
+			minimal: "minimal",
+			low: "low",
+			medium: "medium",
+			high: "high",
+			xhigh: "xhigh",
 			max: null,},
 	},
 	{
@@ -37,13 +37,13 @@ export const OPENCODE_MODELS: ModelDef[] = [
 		contextWindow: 1_048_576,
 		maxTokens: 131_072,
 		input: ["text", "image"],
-		thinkingLevelMap: {
-			off: null,
-			minimal: "low",
-			low: "low",
-			medium: "medium",
-			high: "high",
-			xhigh: "high",
+		thinkingLevelMap: {
+			off: null,
+			minimal: "low",
+			low: "low",
+			medium: "medium",
+			high: "high",
+			xhigh: "high",
 			max: null,},
 	},
 	{
@@ -51,15 +51,15 @@ export const OPENCODE_MODELS: ModelDef[] = [
 		name: "Hy3 (262K)",
 		reasoning: true,
 		contextWindow: 262_144,
-		maxTokens: 262_144,
+		maxTokens: 128_000,
 		input: ["text"],
-		thinkingLevelMap: {
-			off: null,
-			minimal: "minimal",
-			low: "low",
-			medium: "medium",
-			high: "high",
-			xhigh: "xhigh",
+		thinkingLevelMap: {
+			off: null,
+			minimal: "minimal",
+			low: "low",
+			medium: "medium",
+			high: "high",
+			xhigh: "xhigh",
 			max: null,},
 	},
 	{
@@ -69,13 +69,13 @@ export const OPENCODE_MODELS: ModelDef[] = [
 		contextWindow: 1_000_000,
 		maxTokens: 128_000,
 		input: ["text"],
-		thinkingLevelMap: {
-			off: null,
-			minimal: "minimal",
-			low: "low",
-			medium: "medium",
-			high: "high",
-			xhigh: "xhigh",
+		thinkingLevelMap: {
+			off: null,
+			minimal: "minimal",
+			low: "low",
+			medium: "medium",
+			high: "high",
+			xhigh: "xhigh",
 			max: null,},
 	},
 	{
@@ -85,13 +85,13 @@ export const OPENCODE_MODELS: ModelDef[] = [
 		contextWindow: 1_000_000,
 		maxTokens: 262_144,
 		input: ["text"],
-		thinkingLevelMap: {
-			off: null,
-			minimal: "minimal",
-			low: "low",
-			medium: "medium",
-			high: "high",
-			xhigh: "xhigh",
+		thinkingLevelMap: {
+			off: null,
+			minimal: "minimal",
+			low: "low",
+			medium: "medium",
+			high: "high",
+			xhigh: "xhigh",
 			max: null,},
 	},
 	{
@@ -101,37 +101,49 @@ export const OPENCODE_MODELS: ModelDef[] = [
 		contextWindow: 200_000,
 		maxTokens: 32_000,
 		input: ["text"],
-		thinkingLevelMap: {
-			off: null,
-			minimal: null,
-			low: null,
-			medium: null,
-			high: "high",
-			xhigh: null,
+		thinkingLevelMap: {
+			off: null,
+			minimal: null,
+			low: null,
+			medium: null,
+			high: "high",
+			xhigh: null,
 			max: "max",},
 	},
 	{
 		id: "laguna-s-2.1-free",
-		name: "Laguna S 2.1 (1M)",
+		name: "Laguna S 2.1 (256K)",
 		reasoning: true,
-		contextWindow: 1_048_576,
-		maxTokens: 131_072,
+		contextWindow: 262_144,
+		maxTokens: 32_768,
 		input: ["text"],
-		thinkingLevelMap: {
-			off: null,
-			minimal: "minimal",
-			low: "low",
-			medium: "medium",
-			high: "high",
-			xhigh: "xhigh",
+		thinkingLevelMap: {
+			off: null,
+			minimal: "minimal",
+			low: "low",
+			medium: "medium",
+			high: "high",
+			xhigh: "xhigh",
 			max: null,},
 	},
 ];
 
 /**
- * Backward compatibility alias for OPENCODE_MODELS
+ * Shared effort map for Kilo reasoning models — verified live 2026-08-29:
+ * gateway accepts flat reasoning_effort minimal..xhigh for every reasoning
+ * model; stepfun/step-3.7-flash measured monotonic 77→313 thinking chars
+ * across minimal→xhigh. Declaring the map locks the picker (instead of
+ * host guessing) and matches the OpenCode-model pattern.
  */
-export const KNOWN_MODELS = OPENCODE_MODELS;
+const KILO_REASONING_MAP: ThinkingLevelMap = {
+	off: null,
+	minimal: "minimal",
+	low: "low",
+	medium: "medium",
+	high: "high",
+	xhigh: "xhigh",
+	max: null,
+};
 
 /**
  * KiloCode Gateway free models (keyless — https://kilo.ai/docs/gateway).
@@ -146,6 +158,7 @@ export const KILO_MODELS: ModelDef[] = [
 		maxTokens: 512_000,
 		input: ["text", "image"],
 		thinkingFormat: "openrouter",
+		thinkingLevelMap: KILO_REASONING_MAP,
 	},
 	{
 		id: "stepfun/step-3.7-flash:free",
@@ -155,24 +168,27 @@ export const KILO_MODELS: ModelDef[] = [
 		maxTokens: 262_144,
 		input: ["text", "image"],
 		thinkingFormat: "openrouter",
+		thinkingLevelMap: KILO_REASONING_MAP,
 	},
 	{
 		id: "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free",
 		name: "Nemotron 3 Nano Omni",
 		reasoning: true,
 		contextWindow: 256_000,
-		maxTokens: 65_536,
+		maxTokens: 131_072,
 		input: ["text", "image"],
 		thinkingFormat: "openrouter",
+		thinkingLevelMap: KILO_REASONING_MAP,
 	},
 	{
 		id: "nvidia/nemotron-3-ultra-550b-a55b:free",
 		name: "Nemotron 3 Ultra 550B (1M)",
 		reasoning: true,
 		contextWindow: 1_000_000,
-		maxTokens: 65_536,
+		maxTokens: 128_000,
 		input: ["text"],
 		thinkingFormat: "openrouter",
+		thinkingLevelMap: KILO_REASONING_MAP,
 	},
 	{
 		id: "nvidia/nemotron-3.5-lightning:free",
@@ -182,6 +198,7 @@ export const KILO_MODELS: ModelDef[] = [
 		maxTokens: 262_144,
 		input: ["text"],
 		thinkingFormat: "openrouter",
+		thinkingLevelMap: KILO_REASONING_MAP,
 	},
 	{
 		id: "nvidia/nemotron-3-super-120b-a12b:free",
@@ -191,15 +208,17 @@ export const KILO_MODELS: ModelDef[] = [
 		maxTokens: 262_144,
 		input: ["text"],
 		thinkingFormat: "openrouter",
+		thinkingLevelMap: KILO_REASONING_MAP,
 	},
 	{
 		id: "tencent/hy3:free",
 		name: "Tencent Hy3 (Kilo)",
 		reasoning: true,
 		contextWindow: 262_144,
-		maxTokens: 262_144,
+		maxTokens: 128_000,
 		input: ["text"],
 		thinkingFormat: "openrouter",
+		thinkingLevelMap: KILO_REASONING_MAP,
 	},
 	{
 		id: "cohere/north-mini-code:free",
@@ -209,15 +228,17 @@ export const KILO_MODELS: ModelDef[] = [
 		maxTokens: 64_000,
 		input: ["text"],
 		thinkingFormat: "openrouter",
+		thinkingLevelMap: KILO_REASONING_MAP,
 	},
 	{
 		id: "poolside/laguna-s-2.1:free",
 		name: "Laguna S 2.1 (Kilo)",
 		reasoning: true,
-		contextWindow: 1_048_576,
-		maxTokens: 131_072,
+		contextWindow: 262_144,
+		maxTokens: 32_768,
 		input: ["text"],
 		thinkingFormat: "openrouter",
+		thinkingLevelMap: KILO_REASONING_MAP,
 	},
 	{
 		id: "poolside/laguna-xs-2.1:free",
@@ -227,31 +248,37 @@ export const KILO_MODELS: ModelDef[] = [
 		maxTokens: 32_768,
 		input: ["text"],
 		thinkingFormat: "openrouter",
+		thinkingLevelMap: KILO_REASONING_MAP,
 	},
 	{
 		id: "liquid/lfm-2.5-2.6b:free",
 		name: "Liquid LFM 2.5",
 		reasoning: true,
-		contextWindow: 128_000,
+		contextWindow: 65_536,
 		maxTokens: 32_768,
 		input: ["text"],
 		thinkingFormat: "openrouter",
+		thinkingLevelMap: KILO_REASONING_MAP,
 	},
 	{
 		id: "kilo-auto/free",
 		name: "Kilo Auto",
-		reasoning: false,
+		reasoning: true,
 		contextWindow: 256_000,
 		maxTokens: 10_000,
 		input: ["text"],
+		thinkingFormat: "openrouter",
+		thinkingLevelMap: KILO_REASONING_MAP,
 	},
 	{
 		id: "openrouter/free",
 		name: "OpenRouter Auto",
-		reasoning: false,
+		reasoning: true,
 		contextWindow: 200_000,
 		maxTokens: 65_536,
 		input: ["text", "image"],
+		thinkingFormat: "openrouter",
+		thinkingLevelMap: KILO_REASONING_MAP,
 	},
 	{
 		id: "nvidia/nemotron-3.5-content-safety:free",
@@ -261,6 +288,46 @@ export const KILO_MODELS: ModelDef[] = [
 		maxTokens: 8_192,
 		input: ["text", "image"],
 	},
+{
+	id: "meituan/longcat-2.0-free",
+	name: "LongCat 2.0 (1M)",
+	reasoning: true,
+	contextWindow: 1_048_756,
+	maxTokens: 262_144,
+	input: ["text"],
+	thinkingFormat: "openrouter",
+	thinkingLevelMap: KILO_REASONING_MAP,
+},
+{
+	id: "minimax/minimax-m2.7:free",
+	name: "MiniMax M2.7 (free)",
+	reasoning: true,
+	contextWindow: 196_608,
+	maxTokens: 196_608,
+	input: ["text"],
+	thinkingFormat: "openrouter",
+	thinkingLevelMap: KILO_REASONING_MAP,
+},
+{
+	id: "minimax/minimax-m3:free",
+	name: "MiniMax M3 (free)",
+	reasoning: true,
+	contextWindow: 1_048_576,
+	maxTokens: 524_288,
+	input: ["text"],
+	thinkingFormat: "openrouter",
+	thinkingLevelMap: KILO_REASONING_MAP,
+},
+{
+	id: "inclusionai/ling-3.0-flash-fin:free",
+	name: "Ling 3.0 Flash Fin",
+	reasoning: true,
+	contextWindow: 262_144,
+	maxTokens: 32_768,
+	input: ["text"],
+	thinkingFormat: "openrouter",
+	thinkingLevelMap: KILO_REASONING_MAP,
+},
 ];
 
 /**
@@ -276,10 +343,15 @@ export const MODEL_ALIASES: Record<string, string> = {
 	"step-3.7-flash": "stepfun/step-3.7-flash:free",
 	"nemotron-3-nano-omni": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free",
 	"nemotron-3-ultra-550b": "nvidia/nemotron-3-ultra-550b-a55b:free",
+	"nemotron-3.5-lightning": "nvidia/nemotron-3.5-lightning:free",
 	"nemotron-3-super": "nvidia/nemotron-3-super-120b-a12b:free",
 	"north-mini-code": "cohere/north-mini-code:free",
 	"lfm-2.5": "liquid/lfm-2.5-2.6b:free",
 	"content-safety": "nvidia/nemotron-3.5-content-safety:free",
+	"longcat-2.0": "meituan/longcat-2.0-free",
+	"minimax-m2.7": "minimax/minimax-m2.7:free",
+	"minimax-m3": "minimax/minimax-m3:free",
+	"ling-3.0-flash-fin": "inclusionai/ling-3.0-flash-fin:free",
 	// provider-prefixed short aliases (slash-normalized)
 	"hy3:free": "tencent/hy3:free",
 	"laguna-s-2.1:free": "poolside/laguna-s-2.1:free",
@@ -307,7 +379,7 @@ export const KILO_MODEL_IDS = new Set<string>([
 ]);
 
 /**
- * Combined list of all 21 static free models (canonical)
+ * Combined list of all 25 static free models (canonical)
  */
 export const ALL_MODELS: ModelDef[] = [...OPENCODE_MODELS, ...KILO_MODELS];
 

package/src/probe.ts

@@ -0,0 +1,49 @@
+/**
+ * Relay reachability probe: verifies a deployed relay answers requests
+ * by issuing a minimal round trip through it and timing the response.
+ */
+
+export interface RelayProbeResult {
+	ok: boolean;
+	status: number;
+	latencyMs: number;
+	error?: string;
+}
+
+/**
+ * Probe a relay endpoint to confirm it is reachable.
+ *
+ * Sends a minimal request to `<url>/v1/models` carrying the x-relay-target /
+ * x-relay-path headers the relay expects, timing the round trip. Never
+ * throws: network failures, timeouts, and non-2xx statuses are all reported
+ * on the returned result instead.
+ *
+ * The path mirrors the proxy's real relay contract (`relayFetch` in relay.ts):
+ * `x-relay-target` is the upstream origin and `x-relay-path` the full path
+ * including the `/zen` prefix — the workers only allow the exact origins
+ * `https://opencode.ai` / `https://api.kilo.ai` and forward `target + path`.
+ * @param auth Optional per-relay shared secret; sent as x-relay-auth so
+ *   deployed relays (which enforce it) answer the probe instead of 401.
+ */
+export async function probeRelay(url: string, auth?: string): Promise<RelayProbeResult> {
+	const cleanUrl = url.trim().replace(/\/+$/, "");
+	const start = Date.now();
+	try {
+		const res = await fetch(`${cleanUrl}/v1/models`, {
+			headers: {
+				"x-relay-target": "https://opencode.ai",
+				"x-relay-path": "/zen/v1/models",
+				...(auth ? { "x-relay-auth": auth } : {}),
+			},
+			signal: AbortSignal.timeout(5_000),
+		});
+		return { ok: res.ok, status: res.status, latencyMs: Date.now() - start };
+	} catch (e) {
+		return {
+			ok: false,
+			status: 0,
+			latencyMs: Date.now() - start,
+			error: (e as Error)?.message || String(e),
+		};
+	}
+}