@cerefox/memory 0.9.11 → 0.10.1

package/dist/bin/cerefox.js

@@ -7184,7 +7184,7 @@ var exports_meta = {};
 __export(exports_meta, {
   PKG_VERSION: () => PKG_VERSION
 });
-var PKG_VERSION = "0.9.11";
+var PKG_VERSION = "0.10.1";
 var init_meta = () => {};
 
 // ../../node_modules/.bun/tslib@2.8.1/node_modules/tslib/tslib.js
@@ -24852,20 +24852,31 @@ var init_bundled_docs = __esm(() => {
 });
 
 // ../../_shared/embeddings/index.ts
+function openaiEmbeddingConfig() {
+  const env4 = globalThis.process?.env ?? {};
+  const base = env4.CEREFOX_OPENAI_BASE_URL?.replace(/\/+$/, "");
+  const dims = Number.parseInt(env4.CEREFOX_OPENAI_EMBEDDING_DIMENSIONS ?? "", 10);
+  return {
+    url: base ? `${base}/embeddings` : OPENAI_EMBEDDING_URL,
+    model: env4.CEREFOX_OPENAI_EMBEDDING_MODEL || OPENAI_MODEL,
+    dimensions: Number.isNaN(dims) || dims <= 0 ? EMBEDDING_DIMENSIONS : dims
+  };
+}
 async function getEmbedding(text, apiKey) {
   let lastError = null;
+  const cfg = openaiEmbeddingConfig();
   for (let attempt = 0;attempt < EMBEDDING_MAX_RETRIES; attempt++) {
     try {
-      const response = await fetch(OPENAI_EMBEDDING_URL, {
+      const response = await fetch(cfg.url, {
         method: "POST",
         headers: {
           Authorization: `Bearer ${apiKey}`,
           "Content-Type": "application/json"
         },
         body: JSON.stringify({
-          model: OPENAI_MODEL,
+          model: cfg.model,
           input: text,
-          dimensions: EMBEDDING_DIMENSIONS
+          dimensions: cfg.dimensions
         })
       });
       if (!response.ok) {
@@ -24896,18 +24907,19 @@ async function getEmbedding(text, apiKey) {
 }
 async function embedBatchSingleCall(texts, apiKey) {
   let lastError = null;
+  const cfg = openaiEmbeddingConfig();
   for (let attempt = 0;attempt < EMBEDDING_MAX_RETRIES; attempt++) {
     try {
-      const response = await fetch(OPENAI_EMBEDDING_URL, {
+      const response = await fetch(cfg.url, {
         method: "POST",
         headers: {
           Authorization: `Bearer ${apiKey}`,
           "Content-Type": "application/json"
         },
         body: JSON.stringify({
-          model: OPENAI_MODEL,
+          model: cfg.model,
           input: texts,
-          dimensions: EMBEDDING_DIMENSIONS
+          dimensions: cfg.dimensions
         })
       });
       if (!response.ok) {
@@ -53721,6 +53733,20 @@ var require_cli_progress = __commonJS((exports, module) => {
 });
 
 // ../../_shared/mcp-tools/_utils.ts
+function getMaxResponseBytes() {
+  const raw = globalThis.process?.env?.CEREFOX_MAX_RESPONSE_BYTES;
+  if (raw === undefined || raw === "")
+    return MAX_RESPONSE_BYTES;
+  const n = Number.parseInt(raw, 10);
+  return Number.isNaN(n) || n <= 0 ? MAX_RESPONSE_BYTES : n;
+}
+function getMinSearchScore() {
+  const raw = globalThis.process?.env?.CEREFOX_MIN_SEARCH_SCORE;
+  if (raw === undefined || raw === "")
+    return DEFAULT_MIN_SEARCH_SCORE;
+  const n = Number.parseFloat(raw);
+  return Number.isNaN(n) || n < 0 || n > 1 ? DEFAULT_MIN_SEARCH_SCORE : n;
+}
 function applyByteBudget(rows, maxBytes) {
   const accepted = [];
   let usedBytes = 0;
@@ -53748,7 +53774,7 @@ function logUsage(supabase, params) {
     p_extra: params.extra ?? {}
   })).catch(() => {});
 }
-var MAX_RESPONSE_BYTES = 200000;
+var MAX_RESPONSE_BYTES = 200000, DEFAULT_MIN_SEARCH_SCORE = 0.5;
 
 // ../../_shared/mcp-tools/audit-log.ts
 async function handler(supabase, args, ctx) {
@@ -54531,7 +54557,8 @@ async function handler8(supabase, args, ctx) {
     if (!projectId)
       throw new Error(`Project not found: ${project_name}`);
   }
-  const max_bytes = include_content ? Math.min(requested_max_bytes ?? MAX_RESPONSE_BYTES, MAX_RESPONSE_BYTES) : null;
+  const ceiling = getMaxResponseBytes();
+  const max_bytes = include_content ? Math.min(requested_max_bytes ?? ceiling, ceiling) : null;
   const params = {
     p_metadata_filter: metadata_filter,
     p_project_id: projectId,
@@ -54624,10 +54651,11 @@ async function handler9(supabase, args, ctx) {
   const match_count = args.match_count ?? 5;
   const mode = args.mode ?? "docs";
   const alpha = args.alpha ?? 0.7;
-  const min_score = args.min_score ?? 0.5;
+  const min_score = args.min_score ?? getMinSearchScore();
   const metadata_filter = args.metadata_filter ?? null;
   const requested_max_bytes = args.max_bytes;
-  const max_bytes = Math.min(requested_max_bytes ?? MAX_RESPONSE_BYTES, MAX_RESPONSE_BYTES);
+  const ceiling = getMaxResponseBytes();
+  const max_bytes = Math.min(requested_max_bytes ?? ceiling, ceiling);
   if (metadata_filter !== null && metadata_filter !== undefined && (typeof metadata_filter !== "object" || Array.isArray(metadata_filter))) {
     throw new McpInvalidParams("metadata_filter must be a JSON object or null");
   }
@@ -68087,7 +68115,7 @@ function utcStamp() {
   return d.getUTCFullYear().toString() + pad(d.getUTCMonth() + 1) + pad(d.getUTCDate()) + "T" + pad(d.getUTCHours()) + pad(d.getUTCMinutes()) + pad(d.getUTCSeconds()) + "Z";
 }
 async function action(options) {
-  const outDir = resolve(expandHome(options.outputDir ?? "~/.cerefox/backups"));
+  const outDir = resolve(expandHome(options.outputDir ?? process.env.CEREFOX_BACKUP_DIR ?? "~/.cerefox/backups"));
   if (!existsSync2(outDir))
     mkdirSync(outDir, { recursive: true });
   const stamp = utcStamp();
@@ -68133,7 +68161,7 @@ async function action(options) {
   }
 }
 function registerBackup(program2) {
-  program2.command("backup").description("Write a JSON snapshot of the knowledge base.").option("-o, --output-dir <dir>", "Snapshot output directory.", "~/.cerefox/backups").option("-l, --label <label>", "Optional suffix added to the filename.").option("--include-versions", "Include archived versions in the snapshot. (v0.5: ignored — current chunks only.)").option("--git", "Commit the snapshot to the output dir as a git checkpoint. (v0.5: ignored.)").action(action);
+  program2.command("backup").description("Write a JSON snapshot of the knowledge base.").option("-o, --output-dir <dir>", "Snapshot output directory (default: CEREFOX_BACKUP_DIR or ~/.cerefox/backups).").option("-l, --label <label>", "Optional suffix added to the filename.").option("--include-versions", "Include archived versions in the snapshot. (v0.5: ignored — current chunks only.)").option("--git", "Commit the snapshot to the output dir as a git checkpoint. (v0.5: ignored.)").action(action);
 }
 
 // src/cli/commands/completion.ts
@@ -73925,7 +73953,7 @@ import { homedir as homedir5 } from "node:os";
 import { join as join8 } from "node:path";
 
 // ../../_shared/ef-meta/index.ts
-var EF_VERSION = "0.9.3";
+var EF_VERSION = "0.10.1";
 
 // src/cli/util/checks.ts
 init_config();
@@ -75049,6 +75077,22 @@ var DEFAULT_PIPELINE_SETTINGS = {
   versionRetentionHours: 48,
   versionCleanupEnabled: true
 };
+function loadPipelineSettings() {
+  const env4 = globalThis.process?.env ?? {};
+  const intMin = (raw, def, min) => {
+    if (raw === undefined || raw === "")
+      return def;
+    const n = Number.parseInt(raw, 10);
+    return Number.isNaN(n) || n < min ? def : n;
+  };
+  const bool = (raw, def) => raw === undefined || raw === "" ? def : !/^(false|0|no|off)$/i.test(raw.trim());
+  return {
+    maxChunkChars: intMin(env4.CEREFOX_MAX_CHUNK_CHARS, DEFAULT_PIPELINE_SETTINGS.maxChunkChars, 1),
+    minChunkChars: intMin(env4.CEREFOX_MIN_CHUNK_CHARS, DEFAULT_PIPELINE_SETTINGS.minChunkChars, 0),
+    versionRetentionHours: intMin(env4.CEREFOX_VERSION_RETENTION_HOURS, DEFAULT_PIPELINE_SETTINGS.versionRetentionHours, 0),
+    versionCleanupEnabled: bool(env4.CEREFOX_VERSION_CLEANUP_ENABLED, DEFAULT_PIPELINE_SETTINGS.versionCleanupEnabled)
+  };
+}
 
 // src/ingestion/pipeline.ts
 class IngestionPipeline {
@@ -75060,7 +75104,7 @@ class IngestionPipeline {
     this.db = new IngestionDbBridge(deps.supabase);
     this.apiKey = deps.openAiApiKey;
     this.embedderModel = deps.embedderModel ?? "text-embedding-3-small";
-    this.settings = { ...DEFAULT_PIPELINE_SETTINGS, ...deps.settings ?? {} };
+    this.settings = { ...loadPipelineSettings(), ...deps.settings ?? {} };
   }
   async ingestText(opts) {
     const {
@@ -76474,7 +76518,7 @@ async function action28(query, options) {
   }
   const matchCount = parsePositiveInt(options.matchCount, "--match-count", 5);
   const alpha = parseFloat01(options.alpha, "--alpha", 0.7);
-  const minScore = parseFloat01(options.minScore, "--min-score", 0.5);
+  const minScore = parseFloat01(options.minScore, "--min-score", getMinSearchScore());
   const maxBytes = parseNonNegativeInt(options.maxBytes, "--max-bytes", 200000);
   const mode = options.mode ?? "docs";
   if (!["docs", "hybrid", "fts"].includes(mode)) {
@@ -76620,7 +76664,7 @@ async function action28(query, options) {
   }
 }
 function registerSearch(program2) {
-  program2.command("search").description("Search the knowledge base (hybrid FTS + semantic).").argument("<query>", "Natural-language search query.").option("-c, --match-count <n>", "Maximum number of documents to return.", "5").option("-p, --project-name <name>", "Filter results to a specific project.").option("-f, --metadata-filter <json>", "JSON containment filter; only docs whose metadata contains ALL pairs are returned.").option("--mode <mode>", "Search mode: docs (default), hybrid, fts.", "docs").option("--alpha <float>", "Semantic weight 0..1 (default: 0.7).", "0.7").option("--min-score <float>", "Minimum cosine similarity threshold.", "0.5").option("--max-bytes <n>", "Response size budget in bytes.", "200000").option("-r, --requestor <name>", "Agent / user name (recorded in usage log).").option("--json", "Emit machine-readable JSON instead of the default text.").option("--only-metadata", "List matching docs (id, score, chunks, chars, partial/full) WITHOUT their content — like the web UI's collapsed result list. Grab a [id:…] then `cerefox document get <id>`.").action(action28);
+  program2.command("search").description("Search the knowledge base (hybrid FTS + semantic).").argument("<query>", "Natural-language search query.").option("-c, --match-count <n>", "Maximum number of documents to return.", "5").option("-p, --project-name <name>", "Filter results to a specific project.").option("-f, --metadata-filter <json>", "JSON containment filter; only docs whose metadata contains ALL pairs are returned.").option("--mode <mode>", "Search mode: docs (default), hybrid, fts.", "docs").option("--alpha <float>", "Semantic weight 0..1 (default: 0.7).", "0.7").option("--min-score <float>", "Minimum cosine similarity threshold (default: CEREFOX_MIN_SEARCH_SCORE or 0.5).").option("--max-bytes <n>", "Response size budget in bytes.", "200000").option("-r, --requestor <name>", "Agent / user name (recorded in usage log).").option("--json", "Emit machine-readable JSON instead of the default text.").option("--only-metadata", "List matching docs (id, score, chunks, chars, partial/full) WITHOUT their content — like the web UI's collapsed result list. Grab a [id:…] then `cerefox document get <id>`.").action(action28);
 }
 
 // src/cli/commands/self-update.ts
@@ -80241,7 +80285,7 @@ async function runSearch(ctx, opts) {
       p_alpha: 0.7,
       p_use_upgrade: false,
       p_project_id: projectId,
-      p_min_score: 0
+      p_min_score: getMinSearchScore()
     };
     if (metadataFilter)
       params2.p_metadata_filter = metadataFilter;
@@ -81323,6 +81367,51 @@ function registerMetaRoutes(app, ctx) {
   });
 }
 
+// src/web/routes/postgrest-proxy.ts
+var STRIP_RESPONSE_HEADERS = [
+  "connection",
+  "keep-alive",
+  "transfer-encoding",
+  "content-length",
+  "content-encoding",
+  "te",
+  "trailer",
+  "upgrade",
+  "proxy-authenticate",
+  "proxy-authorization"
+];
+function registerPostgrestProxy(app) {
+  const upstream = (process.env.CEREFOX_POSTGREST_UPSTREAM ?? "").trim().replace(/\/+$/, "");
+  if (!upstream)
+    return;
+  app.all("/rest/v1/*", async (c2) => {
+    const url = new URL(c2.req.url);
+    const path = url.pathname.replace(/^\/rest\/v1/, "");
+    const target = `${upstream}${path}${url.search}`;
+    const headers = new Headers(c2.req.raw.headers);
+    headers.delete("host");
+    headers.delete("accept-encoding");
+    const method = c2.req.method;
+    const init = { method, headers };
+    if (method !== "GET" && method !== "HEAD") {
+      init.body = c2.req.raw.body;
+      init.duplex = "half";
+    }
+    let resp;
+    try {
+      resp = await fetch(target, init);
+    } catch (err) {
+      return c2.json({
+        detail: `PostgREST upstream unreachable: ${err instanceof Error ? err.message : String(err)}`
+      }, 502);
+    }
+    const respHeaders = new Headers(resp.headers);
+    for (const h of STRIP_RESPONSE_HEADERS)
+      respHeaders.delete(h);
+    return new Response(resp.body, { status: resp.status, headers: respHeaders });
+  });
+}
+
 // src/web/routes/preferences.ts
 init_config();
 import { existsSync as existsSync14, mkdirSync as mkdirSync4, readFileSync as readFileSync14, writeFileSync as writeFileSync5 } from "node:fs";
@@ -81520,6 +81609,7 @@ function buildApp(ctx = buildWebContext()) {
     app.post("/api/v1/ingest/file", ingest503);
     app.post("/api/v1/documents/:document_id/upload", ingest503);
   }
+  registerPostgrestProxy(app);
   const staticDir = resolveStaticDir();
   if (staticDir) {
     app.use("/static/*", serveStatic({
@@ -81893,7 +81983,10 @@ function registerRenameHusks(program2) {
   }
 }
 function buildProgram() {
-  const program2 = new Command("cerefox").description("Cerefox — user-owned shared memory for AI agents.").version(PKG_VERSION, "-v, --version", "Print the cerefox version and exit.").addOption(new Option("--json", "Emit machine-readable JSON on stdout instead of the default human text. " + "Available on read commands; ignored on commands without a JSON shape.").hideHelp()).showHelpAfterError("(run `cerefox --help` for usage)").enablePositionalOptions().addHelpText("after", "\nResource groups (run `cerefox <group> --help`):\n" + `  document   get · list · edit · delete · restore · ingest · ingest-dir · version {list·archive·unarchive}
+  const progName = process.env.CEREFOX_PROG_NAME || "cerefox";
+  const program2 = new Command(progName).description("Cerefox — user-owned shared memory for AI agents.").version(PKG_VERSION, "-v, --version", `Print the ${progName} version and exit.`).addOption(new Option("--json", "Emit machine-readable JSON on stdout instead of the default human text. " + "Available on read commands; ignored on commands without a JSON shape.").hideHelp()).showHelpAfterError(`(run \`${progName} --help\` for usage)`).enablePositionalOptions().addHelpText("after", `
+Resource groups (run \`${progName} <group> --help\`):
+` + `  document   get · list · edit · delete · restore · ingest · ingest-dir · version {list·archive·unarchive}
 ` + `  project    list · create · edit · delete
 ` + `  metadata   keys · search
 ` + `  audit      list
@@ -81915,8 +82008,8 @@ Exit codes:
 ` + `  1  user error         3  not found (document / version / project)
 ` + `
 Learn more:
-` + `  cerefox docs --list                     # bundled docs (offline)
-` + `  cerefox doctor                          # diagnose your install
+` + `  ${progName} guides list                  # bundled docs (offline)
+` + `  ${progName} doctor                       # diagnose your install
 ` + `  https://github.com/fstamatelopoulos/cerefox
 `);
   registerSearch(program2);