@lincoln504/pi-research 1.7.7 → 1.7.8

package/agent-skill/pi-research/SKILL.md

@@ -2,7 +2,7 @@
 name: pi-research
 description: "Multi-agent web research for any coding agent. Use for anything requiring current information from the internet — news, trends, product/service/company facts, statistics, people, places, events, APIs, library behavior, or 'what is X / how does X work' questions. Also use to search a local knowledge store of previously researched findings before doing live research. Powered by the pi-research engine (search, stealth scraping, security databases, Stack Exchange). Not for reading local files, running project commands, or analyzing this repository's own source."
 license: MIT
-metadata: { "author": "Lincoln Deen", "version": "1.7.7", "package": "@lincoln504/pi-research", "engine": "pi-research-sdk", "openclaw": { "requires": { "bins": ["node"] } } }
+metadata: { "author": "Lincoln Deen", "version": "1.7.8", "package": "@lincoln504/pi-research", "engine": "pi-research-sdk", "openclaw": { "requires": { "bins": ["node"] } } }
 allowed-tools: Bash(node:*)
 ---
 

package/dist/cli-bin.mjs

@@ -37,7 +37,13 @@
 import { existsSync } from 'node:fs';
 import { fileURLToPath } from 'node:url';
 
-const REQUIRED_PEERS = ['@earendil-works/pi-ai', '@earendil-works/pi-coding-agent'];
+const REQUIRED_PEERS = [
+  '@earendil-works/pi-ai',
+  '@earendil-works/pi-coding-agent',
+  // dist/cli.mjs imports it directly (TUI width helpers), so an install that skips
+  // peers can lack it even though pi-coding-agent normally brings it in.
+  '@earendil-works/pi-tui',
+];
 
 /**
  * Probe one peer dependency.

package/dist/cli.mjs

@@ -230,7 +230,12 @@ function redactSecrets(message) {
   out = out.replace(BASIC_AUTH_PATTERN, "Basic [REDACTED]");
   out = out.replace(BEARER_AUTH_PATTERN, "Bearer [REDACTED]");
   out = out.replace(COOKIE_HEADER_PATTERN, (_m, key, sep3) => `${key}${sep3}[REDACTED]`);
-  out = out.replace(SENSITIVE_KV_PATTERN, (_m, key, sep3) => `${key}${sep3}[REDACTED]`);
+  out = out.replace(AUTH_SCHEME_PATTERN, (_m, key, sep3, scheme) => `${key}${sep3}${scheme} [REDACTED]`);
+  out = out.replace(SENSITIVE_KV_PATTERN, (_m, key, sep3, val) => {
+    const q = val[0] === '"' || val[0] === "'" ? val[0] : "";
+    const closed = q !== "" && val.length > 1 && val.endsWith(q);
+    return `${key}${sep3}${q}[REDACTED]${closed ? q : ""}`;
+  });
   out = out.replace(KNOWN_TOKEN_PATTERN, "[REDACTED]");
   out = out.replace(PROVIDER_TOKEN_PATTERN, "[REDACTED]");
   out = out.replace(LONG_HEX_SECRET_PATTERN, "[REDACTED]");
@@ -249,7 +254,7 @@ function redactSecrets(message) {
 function stripTerminalEscapes(message) {
   return message.replace(TERMINAL_ESCAPE_PATTERN, "").replace(CONTROL_CHARS_PATTERN, " ");
 }
-var logContextStorage, MAX_LOG_MESSAGE_LENGTH, ANSI_PATTERN, URL_CREDENTIALS_PATTERN, B_LEFT, B_RIGHT, COOKIE_HEADER_PATTERN, SENSITIVE_KV_PATTERN, KNOWN_TOKEN_PATTERN, JWT_PATTERN, BASIC_AUTH_PATTERN, BEARER_AUTH_PATTERN, PROVIDER_TOKEN_PATTERN, LONG_HEX_SECRET_PATTERN, CONTROL_CHARS_PATTERN, REDACT_SCAN_LENGTH, TOKEN_CHAR_PATTERN, REDACT_SCAN_EXTENSION_MAX, TERMINAL_ESCAPE_PATTERN;
+var logContextStorage, MAX_LOG_MESSAGE_LENGTH, ANSI_PATTERN, URL_CREDENTIALS_PATTERN, B_LEFT, B_RIGHT, COOKIE_HEADER_PATTERN, SENSITIVE_KV_PATTERN, AUTH_SCHEME_PATTERN, KNOWN_TOKEN_PATTERN, JWT_PATTERN, BASIC_AUTH_PATTERN, BEARER_AUTH_PATTERN, PROVIDER_TOKEN_PATTERN, LONG_HEX_SECRET_PATTERN, CONTROL_CHARS_PATTERN, REDACT_SCAN_LENGTH, TOKEN_CHAR_PATTERN, REDACT_SCAN_EXTENSION_MAX, TERMINAL_ESCAPE_PATTERN;
 var init_log_utils = __esm({
   "src/utils/log-utils.ts"() {
     "use strict";
@@ -264,7 +269,11 @@ var init_log_utils = __esm({
       "gi"
     );
     SENSITIVE_KV_PATTERN = new RegExp(
-      `${B_LEFT}(api[_-]?key|apikey|access[_-]?token|auth[_-]?token|refresh[_-]?token|secret|client[_-]?secret|password|passwd|pwd|authorization|bearer|set[_-]?cookie|cookie|session[_-]?id|session|csrf[_-]?token|xsrf[_-]?token|private[_-]?key|key)${B_RIGHT}(["']?\\s*[:=]\\s*["']?)([^\\s"',&)]+)`,
+      `${B_LEFT}(api[_-]?key|apikey|access[_-]?token|auth[_-]?token|refresh[_-]?token|secret|client[_-]?secret|password|passwd|pwd|authorization|bearer|set[_-]?cookie|cookie|session[_-]?id|session|csrf[_-]?token|xsrf[_-]?token|private[_-]?key|token|key)${B_RIGHT}(["']?\\s*[:=]\\s*)("[^"\\r\\n]{0,1024}"|'[^'\\r\\n]{0,1024}'|["']?[^\\s"',&)]+)`,
+      "gi"
+    );
+    AUTH_SCHEME_PATTERN = new RegExp(
+      `${B_LEFT}(authorization)${B_RIGHT}(["']?\\s*[:=]\\s*["']?)([A-Za-z][A-Za-z0-9_-]{0,31})[ \\t]+[^\\s"',&)]+`,
       "gi"
     );
     KNOWN_TOKEN_PATTERN = new RegExp(
@@ -7959,7 +7968,7 @@ var init_store = __esm({
         if (this.db) return;
         try {
           if (!fs13.existsSync(this.options.dbDir)) {
-            fs13.mkdirSync(this.options.dbDir, { recursive: true });
+            fs13.mkdirSync(this.options.dbDir, { recursive: true, mode: 448 });
           }
           this.db = await (await getLancedb()).connect(this.options.dbDir);
           const tableNames = await this.db.tableNames();
@@ -19003,10 +19012,29 @@ ${kept.join("\n")}
 ${rest}` : rest;
 }
 function cleanConvertedMarkdown(markdown) {
-  return stripEmptyFragmentLinks(stripImageLinks(trimFrontMatter(markdown))).replace(/\n{3,}/g, "\n\n").trim();
+  return collapseBlankLines(stripEmptyFragmentLinks(stripImageLinks(trimFrontMatter(markdown)))).trim();
+}
+var CODE_FENCE = /^ {0,3}(`{3,}|~{3,})/;
+function collapseBlankLines(markdown) {
+  if (!markdown.includes("\n\n\n")) return markdown;
+  const out = [];
+  let fence = null;
+  for (const line of markdown.split("\n")) {
+    const m = CODE_FENCE.exec(line);
+    if (fence === null) {
+      if (m) fence = m[1];
+    } else if (m && m[1][0] === fence[0] && m[1].length >= fence.length && line.slice(m[0].length).trim() === "") {
+      fence = null;
+    }
+    if (fence === null && line === "" && out.length > 0 && out[out.length - 1] === "") continue;
+    out.push(line);
+  }
+  return out.join("\n");
 }
 function stripImageLinks(markdown) {
-  return markdown.replace(MARKDOWN_IMAGE_PATTERN, "").replace(IMAGE_LINK_PATTERN, "$1").replace(/\n{3,}/g, "\n\n").trim();
+  return collapseBlankLines(
+    markdown.replace(MARKDOWN_IMAGE_PATTERN, "").replace(IMAGE_LINK_PATTERN, "$1")
+  ).trim();
 }
 function createNativeMarkdownConverter(nativeModule) {
   return async (html) => {
@@ -27194,7 +27222,7 @@ async function exportResearchReport(query, result, _mode, cwd, explicitDir) {
     const filename = `${baseFilename}-${hash}.md`;
     const filepath = join21(targetDir, filename);
     try {
-      await fs15.writeFile(filepath, result, { flag: "wx" });
+      await fs15.writeFile(filepath, result, { flag: "wx", mode: 384 });
       logger.log(`[export] Research report saved to: ${filepath}`);
       return filepath;
     } catch (error) {
@@ -27523,6 +27551,10 @@ function configBlock(det, extraNote) {
   if (extraNote) lines.push(`  \u2022 note:               ${extraNote}`);
   return lines.join("\n");
 }
+function resolveCliDepth(explicit, cfg) {
+  const d = explicit ?? (cfg.QUICK_RESEARCH ? cfg.DEFAULT_RESEARCH_DEPTH : Math.max(1, cfg.DEFAULT_RESEARCH_DEPTH));
+  return d;
+}
 async function cmdResearch(args) {
   const det = await detectCredentials(args.model);
   if (det.problem) {
@@ -27545,7 +27577,7 @@ Error: ${vmsg}
 `);
     return EXIT.USAGE;
   }
-  const depth = args.depth ?? getConfig(process.cwd(), "cli").DEFAULT_RESEARCH_DEPTH;
+  const depth = resolveCliDepth(args.depth, getConfig(process.cwd(), "cli"));
   const runModel = det.model;
   toStderr(`[pi-research] starting research (depth ${depth})${runModel ? ` with ${runModel}` : ""}\u2026
 `);
@@ -28518,5 +28550,6 @@ export {
   exitCodeForSignal,
   makeProgressObserver,
   parseArgs,
-  reportError
+  reportError,
+  resolveCliDepth
 };

package/dist/pdf-extract-worker.mjs

@@ -213,7 +213,11 @@ var COOKIE_HEADER_PATTERN = new RegExp(
   "gi"
 );
 var SENSITIVE_KV_PATTERN = new RegExp(
-  `${B_LEFT}(api[_-]?key|apikey|access[_-]?token|auth[_-]?token|refresh[_-]?token|secret|client[_-]?secret|password|passwd|pwd|authorization|bearer|set[_-]?cookie|cookie|session[_-]?id|session|csrf[_-]?token|xsrf[_-]?token|private[_-]?key|key)${B_RIGHT}(["']?\\s*[:=]\\s*["']?)([^\\s"',&)]+)`,
+  `${B_LEFT}(api[_-]?key|apikey|access[_-]?token|auth[_-]?token|refresh[_-]?token|secret|client[_-]?secret|password|passwd|pwd|authorization|bearer|set[_-]?cookie|cookie|session[_-]?id|session|csrf[_-]?token|xsrf[_-]?token|private[_-]?key|token|key)${B_RIGHT}(["']?\\s*[:=]\\s*)("[^"\\r\\n]{0,1024}"|'[^'\\r\\n]{0,1024}'|["']?[^\\s"',&)]+)`,
+  "gi"
+);
+var AUTH_SCHEME_PATTERN = new RegExp(
+  `${B_LEFT}(authorization)${B_RIGHT}(["']?\\s*[:=]\\s*["']?)([A-Za-z][A-Za-z0-9_-]{0,31})[ \\t]+[^\\s"',&)]+`,
   "gi"
 );
 var KNOWN_TOKEN_PATTERN = new RegExp(
@@ -251,7 +255,12 @@ function redactSecrets(message) {
   out = out.replace(BASIC_AUTH_PATTERN, "Basic [REDACTED]");
   out = out.replace(BEARER_AUTH_PATTERN, "Bearer [REDACTED]");
   out = out.replace(COOKIE_HEADER_PATTERN, (_m, key, sep) => `${key}${sep}[REDACTED]`);
-  out = out.replace(SENSITIVE_KV_PATTERN, (_m, key, sep) => `${key}${sep}[REDACTED]`);
+  out = out.replace(AUTH_SCHEME_PATTERN, (_m, key, sep, scheme) => `${key}${sep}${scheme} [REDACTED]`);
+  out = out.replace(SENSITIVE_KV_PATTERN, (_m, key, sep, val) => {
+    const q = val[0] === '"' || val[0] === "'" ? val[0] : "";
+    const closed = q !== "" && val.length > 1 && val.endsWith(q);
+    return `${key}${sep}${q}[REDACTED]${closed ? q : ""}`;
+  });
   out = out.replace(KNOWN_TOKEN_PATTERN, "[REDACTED]");
   out = out.replace(PROVIDER_TOKEN_PATTERN, "[REDACTED]");
   out = out.replace(LONG_HEX_SECRET_PATTERN, "[REDACTED]");

package/dist/thread-worker.mjs

@@ -60,7 +60,12 @@ function redactSecrets(message) {
   out = out.replace(BASIC_AUTH_PATTERN, "Basic [REDACTED]");
   out = out.replace(BEARER_AUTH_PATTERN, "Bearer [REDACTED]");
   out = out.replace(COOKIE_HEADER_PATTERN, (_m, key, sep) => `${key}${sep}[REDACTED]`);
-  out = out.replace(SENSITIVE_KV_PATTERN, (_m, key, sep) => `${key}${sep}[REDACTED]`);
+  out = out.replace(AUTH_SCHEME_PATTERN, (_m, key, sep, scheme) => `${key}${sep}${scheme} [REDACTED]`);
+  out = out.replace(SENSITIVE_KV_PATTERN, (_m, key, sep, val) => {
+    const q = val[0] === '"' || val[0] === "'" ? val[0] : "";
+    const closed = q !== "" && val.length > 1 && val.endsWith(q);
+    return `${key}${sep}${q}[REDACTED]${closed ? q : ""}`;
+  });
   out = out.replace(KNOWN_TOKEN_PATTERN, "[REDACTED]");
   out = out.replace(PROVIDER_TOKEN_PATTERN, "[REDACTED]");
   out = out.replace(LONG_HEX_SECRET_PATTERN, "[REDACTED]");
@@ -79,7 +84,7 @@ function redactSecrets(message) {
 function stripTerminalEscapes(message) {
   return message.replace(TERMINAL_ESCAPE_PATTERN, "").replace(CONTROL_CHARS_PATTERN, " ");
 }
-var logContextStorage, MAX_LOG_MESSAGE_LENGTH, ANSI_PATTERN, URL_CREDENTIALS_PATTERN, B_LEFT, B_RIGHT, COOKIE_HEADER_PATTERN, SENSITIVE_KV_PATTERN, KNOWN_TOKEN_PATTERN, JWT_PATTERN, BASIC_AUTH_PATTERN, BEARER_AUTH_PATTERN, PROVIDER_TOKEN_PATTERN, LONG_HEX_SECRET_PATTERN, CONTROL_CHARS_PATTERN, REDACT_SCAN_LENGTH, TOKEN_CHAR_PATTERN, REDACT_SCAN_EXTENSION_MAX, TERMINAL_ESCAPE_PATTERN;
+var logContextStorage, MAX_LOG_MESSAGE_LENGTH, ANSI_PATTERN, URL_CREDENTIALS_PATTERN, B_LEFT, B_RIGHT, COOKIE_HEADER_PATTERN, SENSITIVE_KV_PATTERN, AUTH_SCHEME_PATTERN, KNOWN_TOKEN_PATTERN, JWT_PATTERN, BASIC_AUTH_PATTERN, BEARER_AUTH_PATTERN, PROVIDER_TOKEN_PATTERN, LONG_HEX_SECRET_PATTERN, CONTROL_CHARS_PATTERN, REDACT_SCAN_LENGTH, TOKEN_CHAR_PATTERN, REDACT_SCAN_EXTENSION_MAX, TERMINAL_ESCAPE_PATTERN;
 var init_log_utils = __esm({
   "src/utils/log-utils.ts"() {
     "use strict";
@@ -94,7 +99,11 @@ var init_log_utils = __esm({
       "gi"
     );
     SENSITIVE_KV_PATTERN = new RegExp(
-      `${B_LEFT}(api[_-]?key|apikey|access[_-]?token|auth[_-]?token|refresh[_-]?token|secret|client[_-]?secret|password|passwd|pwd|authorization|bearer|set[_-]?cookie|cookie|session[_-]?id|session|csrf[_-]?token|xsrf[_-]?token|private[_-]?key|key)${B_RIGHT}(["']?\\s*[:=]\\s*["']?)([^\\s"',&)]+)`,
+      `${B_LEFT}(api[_-]?key|apikey|access[_-]?token|auth[_-]?token|refresh[_-]?token|secret|client[_-]?secret|password|passwd|pwd|authorization|bearer|set[_-]?cookie|cookie|session[_-]?id|session|csrf[_-]?token|xsrf[_-]?token|private[_-]?key|token|key)${B_RIGHT}(["']?\\s*[:=]\\s*)("[^"\\r\\n]{0,1024}"|'[^'\\r\\n]{0,1024}'|["']?[^\\s"',&)]+)`,
+      "gi"
+    );
+    AUTH_SCHEME_PATTERN = new RegExp(
+      `${B_LEFT}(authorization)${B_RIGHT}(["']?\\s*[:=]\\s*["']?)([A-Za-z][A-Za-z0-9_-]{0,31})[ \\t]+[^\\s"',&)]+`,
       "gi"
     );
     KNOWN_TOKEN_PATTERN = new RegExp(
@@ -1561,6 +1570,7 @@ var init_browser_error_utils = __esm({
 var scraper_utils_exports = {};
 __export(scraper_utils_exports, {
   cleanConvertedMarkdown: () => cleanConvertedMarkdown,
+  collapseBlankLines: () => collapseBlankLines,
   createJsMarkdownConverter: () => createJsMarkdownConverter,
   createNativeMarkdownConverter: () => createNativeMarkdownConverter,
   disposeSsrfSafeFetcher: () => disposeSsrfSafeFetcher,
@@ -1815,10 +1825,28 @@ ${kept.join("\n")}
 ${rest}` : rest;
 }
 function cleanConvertedMarkdown(markdown) {
-  return stripEmptyFragmentLinks(stripImageLinks(trimFrontMatter(markdown))).replace(/\n{3,}/g, "\n\n").trim();
+  return collapseBlankLines(stripEmptyFragmentLinks(stripImageLinks(trimFrontMatter(markdown)))).trim();
+}
+function collapseBlankLines(markdown) {
+  if (!markdown.includes("\n\n\n")) return markdown;
+  const out = [];
+  let fence = null;
+  for (const line of markdown.split("\n")) {
+    const m = CODE_FENCE.exec(line);
+    if (fence === null) {
+      if (m) fence = m[1];
+    } else if (m && m[1][0] === fence[0] && m[1].length >= fence.length && line.slice(m[0].length).trim() === "") {
+      fence = null;
+    }
+    if (fence === null && line === "" && out.length > 0 && out[out.length - 1] === "") continue;
+    out.push(line);
+  }
+  return out.join("\n");
 }
 function stripImageLinks(markdown) {
-  return markdown.replace(MARKDOWN_IMAGE_PATTERN, "").replace(IMAGE_LINK_PATTERN, "$1").replace(/\n{3,}/g, "\n\n").trim();
+  return collapseBlankLines(
+    markdown.replace(MARKDOWN_IMAGE_PATTERN, "").replace(IMAGE_LINK_PATTERN, "$1")
+  ).trim();
 }
 function createNativeMarkdownConverter(nativeModule) {
   return async (html) => {
@@ -1880,7 +1908,7 @@ function validateContent(html, markdown, url) {
     throw error;
   }
 }
-var cachedSsrfFetcher, FRONT_MATTER_SLOTS, MAX_FRONT_MATTER_CHARS, FRONT_MATTER_KEY;
+var cachedSsrfFetcher, FRONT_MATTER_SLOTS, MAX_FRONT_MATTER_CHARS, FRONT_MATTER_KEY, CODE_FENCE;
 var init_scraper_utils = __esm({
   "src/web-research/scraper-utils.ts"() {
     "use strict";
@@ -1907,6 +1935,7 @@ var init_scraper_utils = __esm({
     ];
     MAX_FRONT_MATTER_CHARS = 2e4;
     FRONT_MATTER_KEY = /^([\w.:-]+):(?:\s|$)/;
+    CODE_FENCE = /^ {0,3}(`{3,}|~{3,})/;
   }
 });
 

package/docs/CHANGELOG.md

@@ -5,6 +5,16 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [1.7.8] - 2026-10-01
+
+### Fixed
+
+- **Scraped code blocks keep their blank lines.** `cleanConvertedMarkdown` and `stripImageLinks` collapsed every run of blank lines across the whole document, fenced code included, so a PEP 8 two-blank-line gap, a Makefile or preformatted ASCII was altered in what was stored and cited. The collapse now skips fenced blocks (backtick or tilde, closing fence at least as long as the opening one, an unclosed fence running to the end as in CommonMark) and still tidies prose. Empty-anchor removal still applies inside code on purpose: that is where Quarto/nbdev emit them.
+- **`pi-research research` honours the quick-mode gate for a config-derived depth.** `PI_RESEARCH_DEFAULT_RESEARCH_DEPTH=0` with `QUICK_RESEARCH` off ran quick mode from the CLI, while the documented rule, and the pi tool path, resolve it to 1. An explicit `--depth 0` still runs quick mode, as the skill documents. The rule now lives in one tested function, `resolveCliDepth`.
+- **Log redaction covers more credential shapes.** A bare `token` key (`token=…`, `GITHUB_TOKEN=…`, `?token=…`) was not masked, a quoted value was cut at its first space, comma or ampersand (`password="a b"` left ` b"` in the log), and `Authorization: Token <credential>` (any scheme other than Bearer/Basic) masked only the scheme word. Quoted values are now consumed whole under a 1024-character bound, so the scan stays linear (measured on adversarial 40k inputs: single-digit milliseconds). Plural counters such as `max_tokens=4096` are deliberately not matched.
+- **The `pi-research` bin shim also checks `@earendil-works/pi-tui`.** `dist/cli.mjs` imports it directly, but the preflight only listed `pi-ai` and `pi-coding-agent`, so a peer-skipping install lacking only `pi-tui` still died with a raw `ERR_MODULE_NOT_FOUND`.
+- **Local files that hold research content are no longer group/world readable by default.** The knowledge store directory is created `0700` and exported research reports are written `0600` (previously the process umask decided, typically `0775`/`0644`, and exports can fall back to the shared temp directory). An existing directory is not re-chmodded.
+
 ## [1.7.7] - 2026-10-01
 
 ### Changed

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lincoln504/pi-research",
-  "version": "1.7.7",
+  "version": "1.7.8",
   "description": "Web research for your agents with smart and safe tooling + knowledge store",
   "type": "module",
   "private": false,

package/src/cli-bin.mjs

@@ -37,7 +37,13 @@
 import { existsSync } from 'node:fs';
 import { fileURLToPath } from 'node:url';
 
-const REQUIRED_PEERS = ['@earendil-works/pi-ai', '@earendil-works/pi-coding-agent'];
+const REQUIRED_PEERS = [
+  '@earendil-works/pi-ai',
+  '@earendil-works/pi-coding-agent',
+  // dist/cli.mjs imports it directly (TUI width helpers), so an install that skips
+  // peers can lack it even though pi-coding-agent normally brings it in.
+  '@earendil-works/pi-tui',
+];
 
 /**
  * Probe one peer dependency.

package/src/cli.ts

@@ -580,6 +580,20 @@ interface ResearchArgs {
   json?: boolean;
 }
 
+
+/**
+ * Depth for a CLI research run. An explicit --depth 0 always runs quick mode (the
+ * documented CLI/skill entry). A CONFIG-derived default of 0 only does when
+ * QUICK_RESEARCH is on; otherwise it resolves to 1, as on the pi tool path.
+ */
+export function resolveCliDepth(
+  explicit: number | undefined,
+  cfg: { QUICK_RESEARCH: boolean; DEFAULT_RESEARCH_DEPTH: number },
+): 0 | 1 | 2 | 3 {
+  const d = explicit ?? (cfg.QUICK_RESEARCH ? cfg.DEFAULT_RESEARCH_DEPTH : Math.max(1, cfg.DEFAULT_RESEARCH_DEPTH));
+  return d as 0 | 1 | 2 | 3;
+}
+
 /**
  * Research subcommand. Initializes the SDK, runs deep/quick research, prints the
  * markdown report to stdout, and always shuts down (even on failure).
@@ -607,8 +621,7 @@ export async function cmdResearch(args: ResearchArgs): Promise<number> {
     return EXIT.USAGE;
   }
 
-  const depth: 0 | 1 | 2 | 3 =
-    (args.depth ?? getConfig(process.cwd(), 'cli').DEFAULT_RESEARCH_DEPTH) as 0 | 1 | 2 | 3;
+  const depth = resolveCliDepth(args.depth, getConfig(process.cwd(), 'cli'));
   // det.model already reflects --model-first precedence (detectCredentials above).
   const runModel = det.model;
   toStderr(`[pi-research] starting research (depth ${depth})${runModel ? ` with ${runModel}` : ''}…\n`);

package/src/infrastructure/browser/thread-worker.mjs

@@ -60,7 +60,12 @@ function redactSecrets(message) {
   out = out.replace(BASIC_AUTH_PATTERN, "Basic [REDACTED]");
   out = out.replace(BEARER_AUTH_PATTERN, "Bearer [REDACTED]");
   out = out.replace(COOKIE_HEADER_PATTERN, (_m, key, sep) => `${key}${sep}[REDACTED]`);
-  out = out.replace(SENSITIVE_KV_PATTERN, (_m, key, sep) => `${key}${sep}[REDACTED]`);
+  out = out.replace(AUTH_SCHEME_PATTERN, (_m, key, sep, scheme) => `${key}${sep}${scheme} [REDACTED]`);
+  out = out.replace(SENSITIVE_KV_PATTERN, (_m, key, sep, val) => {
+    const q = val[0] === '"' || val[0] === "'" ? val[0] : "";
+    const closed = q !== "" && val.length > 1 && val.endsWith(q);
+    return `${key}${sep}${q}[REDACTED]${closed ? q : ""}`;
+  });
   out = out.replace(KNOWN_TOKEN_PATTERN, "[REDACTED]");
   out = out.replace(PROVIDER_TOKEN_PATTERN, "[REDACTED]");
   out = out.replace(LONG_HEX_SECRET_PATTERN, "[REDACTED]");
@@ -79,7 +84,7 @@ function redactSecrets(message) {
 function stripTerminalEscapes(message) {
   return message.replace(TERMINAL_ESCAPE_PATTERN, "").replace(CONTROL_CHARS_PATTERN, " ");
 }
-var logContextStorage, MAX_LOG_MESSAGE_LENGTH, ANSI_PATTERN, URL_CREDENTIALS_PATTERN, B_LEFT, B_RIGHT, COOKIE_HEADER_PATTERN, SENSITIVE_KV_PATTERN, KNOWN_TOKEN_PATTERN, JWT_PATTERN, BASIC_AUTH_PATTERN, BEARER_AUTH_PATTERN, PROVIDER_TOKEN_PATTERN, LONG_HEX_SECRET_PATTERN, CONTROL_CHARS_PATTERN, REDACT_SCAN_LENGTH, TOKEN_CHAR_PATTERN, REDACT_SCAN_EXTENSION_MAX, TERMINAL_ESCAPE_PATTERN;
+var logContextStorage, MAX_LOG_MESSAGE_LENGTH, ANSI_PATTERN, URL_CREDENTIALS_PATTERN, B_LEFT, B_RIGHT, COOKIE_HEADER_PATTERN, SENSITIVE_KV_PATTERN, AUTH_SCHEME_PATTERN, KNOWN_TOKEN_PATTERN, JWT_PATTERN, BASIC_AUTH_PATTERN, BEARER_AUTH_PATTERN, PROVIDER_TOKEN_PATTERN, LONG_HEX_SECRET_PATTERN, CONTROL_CHARS_PATTERN, REDACT_SCAN_LENGTH, TOKEN_CHAR_PATTERN, REDACT_SCAN_EXTENSION_MAX, TERMINAL_ESCAPE_PATTERN;
 var init_log_utils = __esm({
   "src/utils/log-utils.ts"() {
     "use strict";
@@ -94,7 +99,11 @@ var init_log_utils = __esm({
       "gi"
     );
     SENSITIVE_KV_PATTERN = new RegExp(
-      `${B_LEFT}(api[_-]?key|apikey|access[_-]?token|auth[_-]?token|refresh[_-]?token|secret|client[_-]?secret|password|passwd|pwd|authorization|bearer|set[_-]?cookie|cookie|session[_-]?id|session|csrf[_-]?token|xsrf[_-]?token|private[_-]?key|key)${B_RIGHT}(["']?\\s*[:=]\\s*["']?)([^\\s"',&)]+)`,
+      `${B_LEFT}(api[_-]?key|apikey|access[_-]?token|auth[_-]?token|refresh[_-]?token|secret|client[_-]?secret|password|passwd|pwd|authorization|bearer|set[_-]?cookie|cookie|session[_-]?id|session|csrf[_-]?token|xsrf[_-]?token|private[_-]?key|token|key)${B_RIGHT}(["']?\\s*[:=]\\s*)("[^"\\r\\n]{0,1024}"|'[^'\\r\\n]{0,1024}'|["']?[^\\s"',&)]+)`,
+      "gi"
+    );
+    AUTH_SCHEME_PATTERN = new RegExp(
+      `${B_LEFT}(authorization)${B_RIGHT}(["']?\\s*[:=]\\s*["']?)([A-Za-z][A-Za-z0-9_-]{0,31})[ \\t]+[^\\s"',&)]+`,
       "gi"
     );
     KNOWN_TOKEN_PATTERN = new RegExp(
@@ -1561,6 +1570,7 @@ var init_browser_error_utils = __esm({
 var scraper_utils_exports = {};
 __export(scraper_utils_exports, {
   cleanConvertedMarkdown: () => cleanConvertedMarkdown,
+  collapseBlankLines: () => collapseBlankLines,
   createJsMarkdownConverter: () => createJsMarkdownConverter,
   createNativeMarkdownConverter: () => createNativeMarkdownConverter,
   disposeSsrfSafeFetcher: () => disposeSsrfSafeFetcher,
@@ -1815,10 +1825,28 @@ ${kept.join("\n")}
 ${rest}` : rest;
 }
 function cleanConvertedMarkdown(markdown) {
-  return stripEmptyFragmentLinks(stripImageLinks(trimFrontMatter(markdown))).replace(/\n{3,}/g, "\n\n").trim();
+  return collapseBlankLines(stripEmptyFragmentLinks(stripImageLinks(trimFrontMatter(markdown)))).trim();
+}
+function collapseBlankLines(markdown) {
+  if (!markdown.includes("\n\n\n")) return markdown;
+  const out = [];
+  let fence = null;
+  for (const line of markdown.split("\n")) {
+    const m = CODE_FENCE.exec(line);
+    if (fence === null) {
+      if (m) fence = m[1];
+    } else if (m && m[1][0] === fence[0] && m[1].length >= fence.length && line.slice(m[0].length).trim() === "") {
+      fence = null;
+    }
+    if (fence === null && line === "" && out.length > 0 && out[out.length - 1] === "") continue;
+    out.push(line);
+  }
+  return out.join("\n");
 }
 function stripImageLinks(markdown) {
-  return markdown.replace(MARKDOWN_IMAGE_PATTERN, "").replace(IMAGE_LINK_PATTERN, "$1").replace(/\n{3,}/g, "\n\n").trim();
+  return collapseBlankLines(
+    markdown.replace(MARKDOWN_IMAGE_PATTERN, "").replace(IMAGE_LINK_PATTERN, "$1")
+  ).trim();
 }
 function createNativeMarkdownConverter(nativeModule) {
   return async (html) => {
@@ -1880,7 +1908,7 @@ function validateContent(html, markdown, url) {
     throw error;
   }
 }
-var cachedSsrfFetcher, FRONT_MATTER_SLOTS, MAX_FRONT_MATTER_CHARS, FRONT_MATTER_KEY;
+var cachedSsrfFetcher, FRONT_MATTER_SLOTS, MAX_FRONT_MATTER_CHARS, FRONT_MATTER_KEY, CODE_FENCE;
 var init_scraper_utils = __esm({
   "src/web-research/scraper-utils.ts"() {
     "use strict";
@@ -1907,6 +1935,7 @@ var init_scraper_utils = __esm({
     ];
     MAX_FRONT_MATTER_CHARS = 2e4;
     FRONT_MATTER_KEY = /^([\w.:-]+):(?:\s|$)/;
+    CODE_FENCE = /^ {0,3}(`{3,}|~{3,})/;
   }
 });
 

package/src/knowledge/store.ts

@@ -279,7 +279,7 @@ export class KnowledgeStore implements IKnowledgeStore {
 
     try {
       if (!fs.existsSync(this.options.dbDir)) {
-        fs.mkdirSync(this.options.dbDir, { recursive: true });
+        fs.mkdirSync(this.options.dbDir, { recursive: true, mode: 0o700 });
       }
       this.db = await (await getLancedb()).connect(this.options.dbDir);
 

package/src/utils/log-utils.ts

@@ -126,8 +126,18 @@ const COOKIE_HEADER_PATTERN = new RegExp(
 // sort key, a map key) is the accepted trade-off — same one already made for
 // the equally generic bare `session`/`cookie` entries below.
 const SENSITIVE_KV_PATTERN = new RegExp(
-  `${B_LEFT}(api[_-]?key|apikey|access[_-]?token|auth[_-]?token|refresh[_-]?token|secret|client[_-]?secret|password|passwd|pwd|authorization|bearer|set[_-]?cookie|cookie|session[_-]?id|session|csrf[_-]?token|xsrf[_-]?token|private[_-]?key|key)${B_RIGHT}` +
-    `(["']?\\s*[:=]\\s*["']?)([^\\s"',&)]+)`,
+  `${B_LEFT}(api[_-]?key|apikey|access[_-]?token|auth[_-]?token|refresh[_-]?token|secret|client[_-]?secret|password|passwd|pwd|authorization|bearer|set[_-]?cookie|cookie|session[_-]?id|session|csrf[_-]?token|xsrf[_-]?token|private[_-]?key|token|key)${B_RIGHT}` +
+    // A quoted value is consumed whole (bounded, so no backtracking blow-up): the
+    // old unquoted class stopped at the first space/comma/&, so `password="a b"`
+    // left ` b"` in the clear. An unterminated quote falls through to that class.
+    `(["']?\\s*[:=]\\s*)("[^"\\r\\n]{0,1024}"|'[^'\\r\\n]{0,1024}'|["']?[^\\s"',&)]+)`,
+  'gi',
+);
+// `Authorization: <scheme> <credential>` for schemes the Bearer/Basic patterns do
+// not know (Token, Digest, ApiKey, ...). The KV pass alone would mask only the
+// scheme word and leave the credential after the space.
+const AUTH_SCHEME_PATTERN = new RegExp(
+  `${B_LEFT}(authorization)${B_RIGHT}(["']?\\s*[:=]\\s*["']?)([A-Za-z][A-Za-z0-9_-]{0,31})[ \\t]+[^\\s"',&)]+`,
   'gi',
 );
 // Well-known opaque credential formats (OpenAI, GitHub, AWS, Slack, Anthropic).
@@ -237,7 +247,12 @@ export function redactSecrets(message: string): string {
   // Whole-header cookie masking before the KV pass, which would otherwise stop
   // at the first whitespace and leave every subsequent pair in the clear.
   out = out.replace(COOKIE_HEADER_PATTERN, (_m, key: string, sep: string) => `${key}${sep}[REDACTED]`);
-  out = out.replace(SENSITIVE_KV_PATTERN, (_m, key: string, sep: string) => `${key}${sep}[REDACTED]`);
+  out = out.replace(AUTH_SCHEME_PATTERN, (_m, key: string, sep: string, scheme: string) => `${key}${sep}${scheme} [REDACTED]`);
+  out = out.replace(SENSITIVE_KV_PATTERN, (_m, key: string, sep: string, val: string) => {
+    const q = val[0] === '"' || val[0] === "'" ? val[0] : '';
+    const closed = q !== '' && val.length > 1 && val.endsWith(q);
+    return `${key}${sep}${q}[REDACTED]${closed ? q : ''}`;
+  });
   out = out.replace(KNOWN_TOKEN_PATTERN, '[REDACTED]');
   out = out.replace(PROVIDER_TOKEN_PATTERN, '[REDACTED]');
   out = out.replace(LONG_HEX_SECRET_PATTERN, '[REDACTED]');

package/src/utils/research-export.ts

@@ -188,7 +188,7 @@ export async function exportResearchReport(
     const filepath = join(targetDir, filename);
 
     try {
-      await fs.writeFile(filepath, result, { flag: 'wx' });
+      await fs.writeFile(filepath, result, { flag: 'wx', mode: 0o600 });
       logger.log(`[export] Research report saved to: ${filepath}`);
       return filepath;
     } catch (error) {

package/src/web-research/pdf-extract-worker.mjs

@@ -213,7 +213,11 @@ var COOKIE_HEADER_PATTERN = new RegExp(
   "gi"
 );
 var SENSITIVE_KV_PATTERN = new RegExp(
-  `${B_LEFT}(api[_-]?key|apikey|access[_-]?token|auth[_-]?token|refresh[_-]?token|secret|client[_-]?secret|password|passwd|pwd|authorization|bearer|set[_-]?cookie|cookie|session[_-]?id|session|csrf[_-]?token|xsrf[_-]?token|private[_-]?key|key)${B_RIGHT}(["']?\\s*[:=]\\s*["']?)([^\\s"',&)]+)`,
+  `${B_LEFT}(api[_-]?key|apikey|access[_-]?token|auth[_-]?token|refresh[_-]?token|secret|client[_-]?secret|password|passwd|pwd|authorization|bearer|set[_-]?cookie|cookie|session[_-]?id|session|csrf[_-]?token|xsrf[_-]?token|private[_-]?key|token|key)${B_RIGHT}(["']?\\s*[:=]\\s*)("[^"\\r\\n]{0,1024}"|'[^'\\r\\n]{0,1024}'|["']?[^\\s"',&)]+)`,
+  "gi"
+);
+var AUTH_SCHEME_PATTERN = new RegExp(
+  `${B_LEFT}(authorization)${B_RIGHT}(["']?\\s*[:=]\\s*["']?)([A-Za-z][A-Za-z0-9_-]{0,31})[ \\t]+[^\\s"',&)]+`,
   "gi"
 );
 var KNOWN_TOKEN_PATTERN = new RegExp(
@@ -251,7 +255,12 @@ function redactSecrets(message) {
   out = out.replace(BASIC_AUTH_PATTERN, "Basic [REDACTED]");
   out = out.replace(BEARER_AUTH_PATTERN, "Bearer [REDACTED]");
   out = out.replace(COOKIE_HEADER_PATTERN, (_m, key, sep) => `${key}${sep}[REDACTED]`);
-  out = out.replace(SENSITIVE_KV_PATTERN, (_m, key, sep) => `${key}${sep}[REDACTED]`);
+  out = out.replace(AUTH_SCHEME_PATTERN, (_m, key, sep, scheme) => `${key}${sep}${scheme} [REDACTED]`);
+  out = out.replace(SENSITIVE_KV_PATTERN, (_m, key, sep, val) => {
+    const q = val[0] === '"' || val[0] === "'" ? val[0] : "";
+    const closed = q !== "" && val.length > 1 && val.endsWith(q);
+    return `${key}${sep}${q}[REDACTED]${closed ? q : ""}`;
+  });
   out = out.replace(KNOWN_TOKEN_PATTERN, "[REDACTED]");
   out = out.replace(PROVIDER_TOKEN_PATTERN, "[REDACTED]");
   out = out.replace(LONG_HEX_SECRET_PATTERN, "[REDACTED]");

package/src/web-research/scraper-utils.ts

@@ -551,20 +551,43 @@ export function trimFrontMatter(markdown: string): string {
  * leaves extra blank lines, so they are collapsed again.
  */
 export function cleanConvertedMarkdown(markdown: string): string {
-  return stripEmptyFragmentLinks(stripImageLinks(trimFrontMatter(markdown)))
-    .replace(/\n{3,}/g, '\n\n')
-    .trim();
+  return collapseBlankLines(stripEmptyFragmentLinks(stripImageLinks(trimFrontMatter(markdown)))).trim();
+}
+
+/** Opening/closing fence of a fenced code block (CommonMark: up to 3 spaces, 3+ backticks or tildes). */
+const CODE_FENCE = /^ {0,3}(`{3,}|~{3,})/;
+
+/**
+ * Collapse runs of blank lines to one, EXCEPT inside fenced code blocks, where
+ * blank lines are content (PEP 8 spacing, Makefiles, ASCII art). An unclosed
+ * fence runs to the end of the document, as in CommonMark.
+ */
+export function collapseBlankLines(markdown: string): string {
+  if (!markdown.includes('\n\n\n')) return markdown;
+  const out: string[] = [];
+  let fence: string | null = null;
+  for (const line of markdown.split('\n')) {
+    const m = CODE_FENCE.exec(line);
+    if (fence === null) {
+      if (m) fence = m[1]!;
+    } else if (m && m[1]![0] === fence[0] && m[1]!.length >= fence.length && line.slice(m[0].length).trim() === '') {
+      fence = null;
+    }
+    if (fence === null && line === '' && out.length > 0 && out[out.length - 1] === '') continue;
+    out.push(line);
+  }
+  return out.join('\n');
 }
 
 /**
  * Strip image links from markdown
  */
 export function stripImageLinks(markdown: string): string {
-  return markdown
-    .replace(MARKDOWN_IMAGE_PATTERN, '')
-    .replace(IMAGE_LINK_PATTERN, '$1')
-    .replace(/\n{3,}/g, '\n\n')
-    .trim();
+  return collapseBlankLines(
+    markdown
+      .replace(MARKDOWN_IMAGE_PATTERN, '')
+      .replace(IMAGE_LINK_PATTERN, '$1'),
+  ).trim();
 }
 
 /**