pi-rehydra 0.0.0-stage → 1.0.2

package/README.en.md

@@ -0,0 +1,60 @@
+# pi-rehydra
+
+Scrub detected secrets from Pi conversations before the main LLM request.
+
+[中文](./README.md)
+
+This extension intercepts the conversation between [Pi](https://pi.dev) and the LLM. Secrets from your `.env` files are replaced with placeholders before they leave your machine, and transparently restored before any tool (shell commands, file writes, etc.) executes locally; a placeholder that cannot be restored blocks that tool call.
+
+It uses Pi's `ExtensionAPI` events: `context_with_system`, `tool_call`, `tool_result`, `message_end`, and `session_start` / `session_shutdown`.
+
+Detection and rehydration are local. The model only sees placeholders; your tools always see real values.
+
+## Install
+
+```bash
+pi install npm:pi-rehydra
+```
+
+Pi writes the package entry to `~/.pi/agent/settings.json`. Edit that file to add extension options:
+
+```json
+{
+  "packages": [
+    {
+      "source": "npm:pi-rehydra"
+    }
+  ],
+  "rehydra": {
+    "envFiles": [".env", ".env.local"],
+    "redactValues": ["sk-live-abc123..."],
+    "allowLiteralTagsIn": ["write", "edit"]
+  }
+}
+```
+
+Every key under `rehydra` is an extension option: `envFiles`, `redactValues`, `minValueLength`, `disableTypes`, `locale`, `vcsIdentities`, `policy`, `anonymizer`, `tagFormat`. `envFiles` defaults to `["**/.env*"]`, skipping `node_modules`, `.git`, and symlinks.
+
+`allowLiteralTagsIn` defaults to an empty array (always block). Tools listed here — typically `write` and `edit` — keep an **unresolvable** literal tag in their arguments instead of aborting, so you can write documentation or tests that intentionally show tag examples. Every allowed occurrence is logged (`allowed literal tag in tool args`, tool name and argument path only). Resolvable tags are still restored in every tool.
+
+## How it works
+
+| Event | Effect |
+|---|---|
+| `session_start` | Instantiate the anonymizer, build the session map, parse the `rehydra` config |
+| `context_with_system` | Anonymize all request messages and inject the rehydra instruction when needed |
+| `tool_call` | Strictly restore PII tags in tool arguments before local execution; an unresolvable tag blocks the call |
+| `tool_result` | Anonymize the completed tool result before it re-enters model context |
+| `message_end` | Rewrite the final assistant message so PII tags are replaced with real values |
+| `session_shutdown` | Dispose of the anonymizer and session registry |
+
+## Limitations
+- Tool-argument rehydration is strict: when a tag is absent from this session's map (or the map is missing, or decryption fails), the tool call is blocked rather than run with `<PII .../>` as a real value. A placeholder can only be restored from the same session's map; the default map is in memory, so tags from before a process restart cannot be restored.
+
+- Model calls that bypass these hooks (rare, e.g. user-initiated notes) are not intercepted.
+- `URL` and `IP_ADDRESS` are disabled by default; re-enable with `disableTypes: []`.
+- This extension cannot guarantee that title or compaction summaries stay sanitized — disable title agents if you need titles to be human-readable.
+
+## License
+
+MIT

package/README.md

@@ -1,3 +1,60 @@
-# Temporary Holding Version
+# pi-rehydra
 
-This version is a temporary placeholder for this package. An operational version to replace this has been submitted for review and is awaiting a staged release.
+在主 LLM 请求发出之前，脱敏 Pi 对话中检测到的密钥。
+
+[English](./README.en.md)
+
+这个扩展拦截 [Pi](https://pi.dev) 与 LLM 之间的对话。你的 `.env` 文件中的密钥在离开本机前被替换为占位符，并在工具（shell 命令、文件写入等）在本地执行前透明还原；无法还原的占位符会阻止该次工具调用。
+
+它依赖 Pi 的 `ExtensionAPI`：`context_with_system`、`tool_call`、`tool_result`、`message_end`、`session_start` / `session_shutdown`。
+
+脱敏和还原都在本地进行；模型只用占位符推理，工具始终拿到真实值。
+
+## 安装
+
+```bash
+pi install npm:pi-rehydra
+```
+
+Pi 会在 `~/.pi/agent/settings.json` 写入包声明。然后编辑该文件添加扩展配置：
+
+```json
+{
+  "packages": [
+    {
+      "source": "npm:pi-rehydra"
+    }
+  ],
+  "rehydra": {
+    "envFiles": [".env", ".env.local"],
+    "redactValues": ["sk-live-abc123..."],
+    "allowLiteralTagsIn": ["write", "edit"]
+  }
+}
+```
+
+`rehydra` 键的所有字段都是扩展配置项：`envFiles`、`redactValues`、`minValueLength`、`disableTypes`、`locale`、`vcsIdentities`、`policy`、`anonymizer`、`tagFormat`。`envFiles` 默认是 `["**/.env*"]`，会跳过 `node_modules`、`.git` 和符号链接。
+
+`allowLiteralTagsIn` 默认是空数组（一律阻断）。列进这个数组的工具（通常是 `write`、`edit`）在参数里出现**无法还原**的字面标签时不再阻断，而是原样保留——用于写文档或测试时故意写标签示例。放行会在日志里留痕（`allowed literal tag in tool args`，只带工具名与参数路径）；能还原的标签在任何工具里都照常还原。
+
+## 工作原理
+
+| 钩子 | 作用 |
+|---|---|
+| `session_start` | 实例化匿名器、装配会话映射、解析 `rehydra` 配置 |
+| `context_with_system` | 匿名化模型请求前的所有消息，并按需追加 rehydra 指令 |
+| `tool_call` | 在工具执行前严格还原工具入参中的 PII 标签；任一标签无法还原即阻断该次调用 |
+| `tool_result` | 在结果返回给模型前脱敏真实 PII |
+| `message_end` | 改写最终回答的文本块，把 PII 标签还原为真实值 |
+| `session_shutdown` | 释放匿名器与会话注册表 |
+
+## 限制
+- 工具入参还原是严格的：标签不在本次会话映射中（或映射缺失、解密失败）时，该次工具调用会被阻止，不会把 `<PII .../>` 当作真实值执行。占位符只能由同一条会话的映射还原；默认使用内存映射，进程重启后旧标签无法还原。
+
+- 不会拦截用户在 UI 里粘贴的、且未走这些 hook 的模型调用。
+- 默认禁用 `URL` 和 `IP_ADDRESS`；通过 `disableTypes: []` 重新启用。
+- Pi 扩展无法保证消息标题（title）与元数据走相同的 hook，因此这些字段仍可能包含占位符。建议在脱敏时启用完整的 PII 脱敏策略。
+
+## License
+
+MIT

package/dist/browser.d.ts

@@ -0,0 +1,54 @@
+/**
+ * Rehydra Module - Browser Entry Point
+ * Browser-safe entry point for on-device PII anonymization
+ *
+ * This module excludes Node.js-specific exports like SQLitePIIStorageProvider.
+ * For full functionality including SQLite, use the main entry point or import from 'rehydra/node'.
+ */
+export * from "./types/index.js";
+export type { Recognizer } from "./recognizers/index.js";
+export { RegexRecognizer, RecognizerRegistry, createDefaultRegistry, createRegistry, getGlobalRegistry, emailRecognizer, phoneRecognizer, ukPostalCodeRecognizer, streetAddressRecognizer, UK_POSTAL_CODE_PATTERN, isValidUKPostalCode, normalizeUKPostalCode, ibanRecognizer, bicSwiftRecognizer, creditCardRecognizer, ipAddressRecognizer, urlRecognizer, createCustomIdRecognizer, createCaseIdRecognizer, createCustomerIdRecognizer, } from "./recognizers/index.js";
+export { NERModel, NERModelStub, createNERModel, createNERModelStub, WordPieceTokenizer, loadVocabFromFile, parseVocab, loadRuntime, detectRuntime, getRuntimeType, type INERModel, type NERModelConfig, type NERPrediction, type NERModelMode, type DownloadProgressCallback, type OrtSessionOptions, MODEL_REGISTRY, getModelCacheDir, isModelDownloaded, downloadModel, ensureModel, clearModelCache, listDownloadedModels, } from "./ner/index.js";
+export { prenormalize, resolveEntities, tagEntities, validateOutput, generateTag, parseTag, rehydrate, enrichSemantics, inferGender, classifyLocation, getDatabaseStats, hasName, hasLocation, isSemanticDataAvailable, isSemanticDataDownloaded, getSemanticDataCacheDir, getDataDirectory, downloadSemanticData, ensureSemanticData, initializeSemanticData, loadSemanticData, clearSemanticData, clearSemanticDataCache, getSemanticDataInfo, SEMANTIC_DATA_FILES, extractTitle, extractTitlesFromSpans, mergeAdjacentTitleSpans, getTitlesForLanguage, getAllTitles, startsWithTitle, isOnlyTitle, type SemanticDataFileInfo, type EnricherConfig, type GenderResult, type LocationResult, type TitleExtractionResult, } from "./pipeline/index.js";
+export type { KeyProvider } from "./crypto/index.js";
+export { encryptPIIMap, decryptPIIMap, generateKey, deriveKey, generateSalt, InMemoryKeyProvider, ConfigKeyProvider, validateKey, secureCompare, uint8ArrayToBase64, base64ToUint8Array, } from "./crypto/index.js";
+export { getStorageProvider, isNode, isBrowser, resetStorageProvider, setStorageProvider, type StorageProvider, } from "#storage";
+export { type PIIStorageProvider, type PIIMapMetadata, type StoredPIIMap, type ListOptions, type AnonymizerSession, InMemoryPIIStorageProvider, IndexedDBPIIStorageProvider, } from "./storage/browser.js";
+export { join as pathJoin, dirname as pathDirname, basename as pathBasename, normalize as pathNormalize, extname as pathExtname, isAbsolute as pathIsAbsolute, } from "./utils/path.js";
+import { Anonymizer as AnonymizerCore, anonymize as anonymizeCore, anonymizeRegexOnly as anonymizeRegexOnlyCore, anonymizeWithNER as anonymizeWithNERCore, type AnonymizerConfig, type NERConfig } from "./core/index.js";
+export type { AnonymizerConfig, NERConfig };
+/**
+ * Anonymizer instance
+ * Main class for performing PII anonymization
+ */
+export declare class Anonymizer extends AnonymizerCore {
+    constructor(config?: AnonymizerConfig);
+}
+/**
+ * Creates an anonymizer with the specified configuration
+ *
+ * @example
+ * ```typescript
+ * // Regex-only (no NER)
+ * const anonymizer = createAnonymizer();
+ *
+ * // With NER (auto-downloads model on first use)
+ * const anonymizer = createAnonymizer({
+ *   ner: { mode: 'quantized' }
+ * });
+ *
+ * // With NER and progress callback
+ * const anonymizer = createAnonymizer({
+ *   ner: {
+ *     mode: 'standard',
+ *     onStatus: (status) => console.log(status),
+ *     onDownloadProgress: (p) => console.log(`${p.file}: ${p.percent}%`)
+ *   }
+ * });
+ * ```
+ */
+export declare function createAnonymizer(config?: AnonymizerConfig): Anonymizer;
+export declare const anonymize: typeof anonymizeCore;
+export declare const anonymizeRegexOnly: typeof anonymizeRegexOnlyCore;
+export declare const anonymizeWithNER: typeof anonymizeWithNERCore;
+//# sourceMappingURL=browser.d.ts.map

package/dist/browser.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"browser.d.ts","sourceRoot":"","sources":["../src/browser.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAGH,cAAc,kBAAkB,CAAC;AAGjC,YAAY,EAAE,UAAU,EAAE,MAAM,wBAAwB,CAAC;AACzD,OAAO,EACL,eAAe,EACf,kBAAkB,EAClB,qBAAqB,EACrB,cAAc,EACd,iBAAiB,EACjB,eAAe,EACf,eAAe,EACf,sBAAsB,EACtB,uBAAuB,EACvB,sBAAsB,EACtB,mBAAmB,EACnB,qBAAqB,EACrB,cAAc,EACd,kBAAkB,EAClB,oBAAoB,EACpB,mBAAmB,EACnB,aAAa,EACb,wBAAwB,EACxB,sBAAsB,EACtB,0BAA0B,GAC3B,MAAM,wBAAwB,CAAC;AAGhC,OAAO,EACL,QAAQ,EACR,YAAY,EACZ,cAAc,EACd,kBAAkB,EAClB,kBAAkB,EAClB,iBAAiB,EACjB,UAAU,EACV,WAAW,EACX,aAAa,EACb,cAAc,EACd,KAAK,SAAS,EACd,KAAK,cAAc,EACnB,KAAK,aAAa,EAClB,KAAK,YAAY,EACjB,KAAK,wBAAwB,EAC7B,KAAK,iBAAiB,EACtB,cAAc,EACd,gBAAgB,EAChB,iBAAiB,EACjB,aAAa,EACb,WAAW,EACX,eAAe,EACf,oBAAoB,GACrB,MAAM,gBAAgB,CAAC;AAGxB,OAAO,EACL,YAAY,EACZ,eAAe,EACf,WAAW,EACX,cAAc,EACd,WAAW,EACX,QAAQ,EACR,SAAS,EACT,eAAe,EACf,WAAW,EACX,gBAAgB,EAChB,gBAAgB,EAChB,OAAO,EACP,WAAW,EAEX,uBAAuB,EACvB,wBAAwB,EACxB,uBAAuB,EACvB,gBAAgB,EAChB,oBAAoB,EACpB,kBAAkB,EAClB,sBAAsB,EACtB,gBAAgB,EAChB,iBAAiB,EACjB,sBAAsB,EACtB,mBAAmB,EACnB,mBAAmB,EAEnB,YAAY,EACZ,sBAAsB,EACtB,uBAAuB,EACvB,oBAAoB,EACpB,YAAY,EACZ,eAAe,EACf,WAAW,EACX,KAAK,oBAAoB,EACzB,KAAK,cAAc,EACnB,KAAK,YAAY,EACjB,KAAK,cAAc,EACnB,KAAK,qBAAqB,GAC3B,MAAM,qBAAqB,CAAC;AAG7B,YAAY,EAAE,WAAW,EAAE,MAAM,mBAAmB,CAAC;AACrD,OAAO,EACL,aAAa,EACb,aAAa,EACb,WAAW,EACX,SAAS,EACT,YAAY,EACZ,mBAAmB,EACnB,iBAAiB,EACjB,WAAW,EACX,aAAa,EACb,kBAAkB,EAClB,kBAAkB,GACnB,MAAM,mBAAmB,CAAC;AAG3B,OAAO,EACL,kBAAkB,EAClB,MAAM,EACN,SAAS,EACT,oBAAoB,EACpB,kBAAkB,EAClB,KAAK,eAAe,GACrB,MAAM,UAAU,CAAC;AAGlB,OAAO,EACL,KAAK,kBAAkB,EACvB,KAAK,cAAc,EACnB,KAAK,YAAY,EACjB,KAAK,WAAW,EAChB,KAAK,iBAAiB,EACtB,0BAA0B,EAC1B,2BAA2B,GAG5B,MAAM,sBAAsB,CAAC;AAG9B,OAAO,EACL,IAAI,IAAI,QAAQ,EAChB,OAAO,IAAI,WAAW,EACtB,QAAQ,IAAI,YAAY,EACxB,SAAS,IAAI,aAAa,EAC1B,OAAO,IAAI,WAAW,EACtB,UAAU,IAAI,cAAc,GAC7B,MAAM,iBAAiB,CAAC;AAGzB,OAAO,EACL,UAAU,IAAI,cAAc,EAC5B,SAAS,IAAI,aAAa,EAC1B,kBAAkB,IAAI,sBAAsB,EAC5C,gBAAgB,IAAI,oBAAoB,EACxC,KAAK,gBAAgB,EACrB,KAAK,SAAS,EACf,MAAM,iBAAiB,CAAC;AAMzB,YAAY,EAAE,gBAAgB,EAAE,SAAS,EAAE,CAAC;AAE5C;;;GAGG;AACH,qBAAa,UAAW,SAAQ,cAAc;gBAChC,MAAM,GAAE,gBAAqB;CAU1C;AAED;;;;;;;;;;;;;;;;;;;;;;GAsBG;AACH,wBAAgB,gBAAgB,CAAC,MAAM,CAAC,EAAE,gBAAgB,GAAG,UAAU,CAEtE;AAGD,eAAO,MAAM,SAAS,sBAAgB,CAAC;AACvC,eAAO,MAAM,kBAAkB,+BAAyB,CAAC;AACzD,eAAO,MAAM,gBAAgB,6BAAuB,CAAC"}

package/dist/browser.js

@@ -0,0 +1,74 @@
+/**
+ * Rehydra Module - Browser Entry Point
+ * Browser-safe entry point for on-device PII anonymization
+ *
+ * This module excludes Node.js-specific exports like SQLitePIIStorageProvider.
+ * For full functionality including SQLite, use the main entry point or import from 'rehydra/node'.
+ */
+// Re-export types
+export * from "./types/index.js";
+export { RegexRecognizer, RecognizerRegistry, createDefaultRegistry, createRegistry, getGlobalRegistry, emailRecognizer, phoneRecognizer, ukPostalCodeRecognizer, streetAddressRecognizer, UK_POSTAL_CODE_PATTERN, isValidUKPostalCode, normalizeUKPostalCode, ibanRecognizer, bicSwiftRecognizer, creditCardRecognizer, ipAddressRecognizer, urlRecognizer, createCustomIdRecognizer, createCaseIdRecognizer, createCustomerIdRecognizer, } from "./recognizers/index.js";
+// Re-export NER components
+export { NERModel, NERModelStub, createNERModel, createNERModelStub, WordPieceTokenizer, loadVocabFromFile, parseVocab, loadRuntime, detectRuntime, getRuntimeType, MODEL_REGISTRY, getModelCacheDir, isModelDownloaded, downloadModel, ensureModel, clearModelCache, listDownloadedModels, } from "./ner/index.js";
+// Re-export pipeline components
+export { prenormalize, resolveEntities, tagEntities, validateOutput, generateTag, parseTag, rehydrate, enrichSemantics, inferGender, classifyLocation, getDatabaseStats, hasName, hasLocation, 
+// Semantic data loader exports
+isSemanticDataAvailable, isSemanticDataDownloaded, getSemanticDataCacheDir, getDataDirectory, downloadSemanticData, ensureSemanticData, initializeSemanticData, loadSemanticData, clearSemanticData, clearSemanticDataCache, getSemanticDataInfo, SEMANTIC_DATA_FILES, 
+// Title extractor exports
+extractTitle, extractTitlesFromSpans, mergeAdjacentTitleSpans, getTitlesForLanguage, getAllTitles, startsWithTitle, isOnlyTitle, } from "./pipeline/index.js";
+export { encryptPIIMap, decryptPIIMap, generateKey, deriveKey, generateSalt, InMemoryKeyProvider, ConfigKeyProvider, validateKey, secureCompare, uint8ArrayToBase64, base64ToUint8Array, } from "./crypto/index.js";
+// Re-export storage utilities (file system abstraction)
+export { getStorageProvider, isNode, isBrowser, resetStorageProvider, setStorageProvider, } from "#storage";
+// Re-export PII storage providers (browser-safe only - no SQLite)
+export { InMemoryPIIStorageProvider, IndexedDBPIIStorageProvider,
+// Note: SQLitePIIStorageProvider is NOT exported in browser build
+// Import from 'rehydra/storage/sqlite' if needed in Node.js
+ } from "./storage/browser.js";
+// Re-export path utilities
+export { join as pathJoin, dirname as pathDirname, basename as pathBasename, normalize as pathNormalize, extname as pathExtname, isAbsolute as pathIsAbsolute, } from "./utils/path.js";
+// Import core anonymizer components
+import { Anonymizer as AnonymizerCore, anonymize as anonymizeCore, anonymizeRegexOnly as anonymizeRegexOnlyCore, anonymizeWithNER as anonymizeWithNERCore, } from "./core/index.js";
+// Import session implementation
+import { AnonymizerSessionImpl } from "./storage/session-base.js";
+/**
+ * Anonymizer instance
+ * Main class for performing PII anonymization
+ */
+export class Anonymizer extends AnonymizerCore {
+    constructor(config = {}) {
+        super(config, (anonymizer, sessionId, storage, keyProvider) => {
+            return new AnonymizerSessionImpl(anonymizer, sessionId, storage, keyProvider);
+        });
+    }
+}
+/**
+ * Creates an anonymizer with the specified configuration
+ *
+ * @example
+ * ```typescript
+ * // Regex-only (no NER)
+ * const anonymizer = createAnonymizer();
+ *
+ * // With NER (auto-downloads model on first use)
+ * const anonymizer = createAnonymizer({
+ *   ner: { mode: 'quantized' }
+ * });
+ *
+ * // With NER and progress callback
+ * const anonymizer = createAnonymizer({
+ *   ner: {
+ *     mode: 'standard',
+ *     onStatus: (status) => console.log(status),
+ *     onDownloadProgress: (p) => console.log(`${p.file}: ${p.percent}%`)
+ *   }
+ * });
+ * ```
+ */
+export function createAnonymizer(config) {
+    return new Anonymizer(config);
+}
+// Re-export standalone functions from core
+export const anonymize = anonymizeCore;
+export const anonymizeRegexOnly = anonymizeRegexOnlyCore;
+export const anonymizeWithNER = anonymizeWithNERCore;
+//# sourceMappingURL=browser.js.map

package/dist/browser.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"browser.js","sourceRoot":"","sources":["../src/browser.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH,kBAAkB;AAClB,cAAc,kBAAkB,CAAC;AAIjC,OAAO,EACL,eAAe,EACf,kBAAkB,EAClB,qBAAqB,EACrB,cAAc,EACd,iBAAiB,EACjB,eAAe,EACf,eAAe,EACf,sBAAsB,EACtB,uBAAuB,EACvB,sBAAsB,EACtB,mBAAmB,EACnB,qBAAqB,EACrB,cAAc,EACd,kBAAkB,EAClB,oBAAoB,EACpB,mBAAmB,EACnB,aAAa,EACb,wBAAwB,EACxB,sBAAsB,EACtB,0BAA0B,GAC3B,MAAM,wBAAwB,CAAC;AAEhC,2BAA2B;AAC3B,OAAO,EACL,QAAQ,EACR,YAAY,EACZ,cAAc,EACd,kBAAkB,EAClB,kBAAkB,EAClB,iBAAiB,EACjB,UAAU,EACV,WAAW,EACX,aAAa,EACb,cAAc,EAOd,cAAc,EACd,gBAAgB,EAChB,iBAAiB,EACjB,aAAa,EACb,WAAW,EACX,eAAe,EACf,oBAAoB,GACrB,MAAM,gBAAgB,CAAC;AAExB,gCAAgC;AAChC,OAAO,EACL,YAAY,EACZ,eAAe,EACf,WAAW,EACX,cAAc,EACd,WAAW,EACX,QAAQ,EACR,SAAS,EACT,eAAe,EACf,WAAW,EACX,gBAAgB,EAChB,gBAAgB,EAChB,OAAO,EACP,WAAW;AACX,+BAA+B;AAC/B,uBAAuB,EACvB,wBAAwB,EACxB,uBAAuB,EACvB,gBAAgB,EAChB,oBAAoB,EACpB,kBAAkB,EAClB,sBAAsB,EACtB,gBAAgB,EAChB,iBAAiB,EACjB,sBAAsB,EACtB,mBAAmB,EACnB,mBAAmB;AACnB,0BAA0B;AAC1B,YAAY,EACZ,sBAAsB,EACtB,uBAAuB,EACvB,oBAAoB,EACpB,YAAY,EACZ,eAAe,EACf,WAAW,GAMZ,MAAM,qBAAqB,CAAC;AAI7B,OAAO,EACL,aAAa,EACb,aAAa,EACb,WAAW,EACX,SAAS,EACT,YAAY,EACZ,mBAAmB,EACnB,iBAAiB,EACjB,WAAW,EACX,aAAa,EACb,kBAAkB,EAClB,kBAAkB,GACnB,MAAM,mBAAmB,CAAC;AAE3B,wDAAwD;AACxD,OAAO,EACL,kBAAkB,EAClB,MAAM,EACN,SAAS,EACT,oBAAoB,EACpB,kBAAkB,GAEnB,MAAM,UAAU,CAAC;AAElB,kEAAkE;AAClE,OAAO,EAML,0BAA0B,EAC1B,2BAA2B;AAC3B,kEAAkE;AAClE,4DAA4D;EAC7D,MAAM,sBAAsB,CAAC;AAE9B,2BAA2B;AAC3B,OAAO,EACL,IAAI,IAAI,QAAQ,EAChB,OAAO,IAAI,WAAW,EACtB,QAAQ,IAAI,YAAY,EACxB,SAAS,IAAI,aAAa,EAC1B,OAAO,IAAI,WAAW,EACtB,UAAU,IAAI,cAAc,GAC7B,MAAM,iBAAiB,CAAC;AAEzB,oCAAoC;AACpC,OAAO,EACL,UAAU,IAAI,cAAc,EAC5B,SAAS,IAAI,aAAa,EAC1B,kBAAkB,IAAI,sBAAsB,EAC5C,gBAAgB,IAAI,oBAAoB,GAGzC,MAAM,iBAAiB,CAAC;AAEzB,gCAAgC;AAChC,OAAO,EAAE,qBAAqB,EAAE,MAAM,2BAA2B,CAAC;AAKlE;;;GAGG;AACH,MAAM,OAAO,UAAW,SAAQ,cAAc;IAC5C,YAAY,SAA2B,EAAE;QACvC,KAAK,CAAC,MAAM,EAAE,CAAC,UAAU,EAAE,SAAS,EAAE,OAAO,EAAE,WAAW,EAAE,EAAE;YAC5D,OAAO,IAAI,qBAAqB,CAC9B,UAAU,EACV,SAAS,EACT,OAAO,EACP,WAAW,CACZ,CAAC;QACJ,CAAC,CAAC,CAAC;IACL,CAAC;CACF;AAED;;;;;;;;;;;;;;;;;;;;;;GAsBG;AACH,MAAM,UAAU,gBAAgB,CAAC,MAAyB;IACxD,OAAO,IAAI,UAAU,CAAC,MAAM,CAAC,CAAC;AAChC,CAAC;AAED,2CAA2C;AAC3C,MAAM,CAAC,MAAM,SAAS,GAAG,aAAa,CAAC;AACvC,MAAM,CAAC,MAAM,kBAAkB,GAAG,sBAAsB,CAAC;AACzD,MAAM,CAAC,MAAM,gBAAgB,GAAG,oBAAoB,CAAC"}

package/dist/core/anonymizer.d.ts

@@ -0,0 +1,309 @@
+/**
+ * Core Anonymizer Module
+ * Shared implementation for both browser and Node.js entry points
+ */
+import { AnonymizationResult, AnonymizationPolicy, AnonymizationMode, SemanticConfig, SecretsConfig, PIIType, TagFormat } from "../types/index.js";
+import { RecognizerRegistry } from "../recognizers/index.js";
+import { type INERModel, type OrtSessionOptions } from "../ner/index.js";
+import { type NERModelMode, type DownloadProgressCallback } from "../ner/model-manager.js";
+import { type RawPIIMap } from "../pipeline/tagger.js";
+import { type KeyProvider } from "../crypto/index.js";
+export type { RawPIIMap } from "../pipeline/tagger.js";
+import type { PIIStorageProvider, AnonymizerSession } from "../storage/types.js";
+export type { PIIStorageProvider, AnonymizerSession };
+/**
+ * NER configuration options
+ */
+export interface NERConfig {
+    /**
+     * NER model mode:
+     * - 'standard': Full-size multilingual model (~1.1 GB)
+     * - 'quantized': Smaller quantized model (~280 MB)
+     * - 'disabled': No NER, regex-only detection
+     * - 'custom': Use custom model paths
+     */
+    mode: NERModelMode;
+    /**
+     * Custom model path (required when mode is 'custom')
+     */
+    modelPath?: string;
+    /**
+     * Custom vocab path (required when mode is 'custom')
+     */
+    vocabPath?: string;
+    /**
+     * Whether to auto-download model if not present
+     * @default true
+     */
+    autoDownload?: boolean;
+    /**
+     * Callback for download progress
+     */
+    onDownloadProgress?: DownloadProgressCallback;
+    /**
+     * Callback for status messages
+     */
+    onStatus?: (status: string) => void;
+    /**
+     * Confidence thresholds per PII type (0.0 - 1.0)
+     * Overrides default thresholds for specified types
+     * @example { PERSON: 0.8, ORG: 0.7 }
+     */
+    thresholds?: Partial<Record<PIIType, number>>;
+    /**
+     * ONNX session options for performance tuning
+     * Allows customizing execution providers, threading, and graph optimizations
+     * @example { executionProviders: ['CoreMLExecutionProvider', 'CPUExecutionProvider'] }
+     */
+    sessionOptions?: OrtSessionOptions;
+    /**
+     * Inference backend: 'local' (default) or 'inference-server'
+     * - 'local': Local ONNX Runtime inference (CPU)
+     * - 'inference-server': Remote GPU inference via HTTP (enterprise deployment)
+     * @default 'local'
+     */
+    backend?: "local" | "inference-server";
+    /**
+     * Inference server URL (required when backend is 'inference-server')
+     * @example 'http://localhost:8080'
+     */
+    inferenceServerUrl?: string;
+    /**
+     * Inference server request timeout in milliseconds (default: 30000)
+     * Only used when backend is 'inference-server'
+     */
+    inferenceServerTimeout?: number;
+    /**
+     * Enable case-insensitive fallback for detecting lowercase names.
+     * Runs a second NER pass on title-cased text and merges new detections.
+     * Doubles NER inference time but catches names like "tom" that the
+     * case-sensitive model would otherwise miss.
+     * @default false
+     */
+    caseFallback?: boolean;
+    /**
+     * Confidence penalty multiplier for case-fallback detections (0.0 - 1.0).
+     * Applied as: confidence * caseFallbackPenalty
+     * @default 0.85
+     */
+    caseFallbackPenalty?: number;
+    /**
+     * Number of content tokens shared by adjacent windows when a long input is
+     * split into multiple inference windows. Larger overlap improves recall for
+     * entities near window boundaries at the cost of more windows.
+     * @default min(64, floor((maxLength - 2) / 4))
+     */
+    windowOverlapTokens?: number;
+    /**
+     * Maximum number of inference windows per input. Inputs that tokenize to
+     * more windows are scanned only up to this bound (onWindowCap fires once per
+     * capped inference pass — so twice for one input when caseFallback runs a
+     * second pass). Guards against unbounded inference cost on very large
+     * opaque payloads (e.g. multi-megabyte base64 attachments).
+     * @default undefined (unbounded — full coverage)
+     */
+    maxWindowsPerInput?: number;
+    /**
+     * Called when an input exceeds maxWindowsPerInput and its tail is left
+     * unscanned by NER. Callers should log/meter this — it is a coverage
+     * reduction, not an error. When absent, a console warning is emitted.
+     * `inputLength` is the full input character count.
+     */
+    onWindowCap?: (info: {
+        scannedWindows: number;
+        inputLength: number;
+    }) => void;
+}
+/**
+ * Anonymizer configuration
+ */
+export interface AnonymizerConfig {
+    /** Recognizer registry (uses default if not provided) */
+    registry?: RecognizerRegistry;
+    /**
+     * Anonymization mode:
+     * - 'pseudonymize': Reversible anonymization with encrypted PII map (default)
+     * - 'anonymize': Irreversible anonymization without PII map
+     * @default 'pseudonymize'
+     */
+    mode?: AnonymizationMode;
+    /**
+     * NER configuration
+     * @default { mode: 'disabled' }
+     */
+    ner?: NERConfig;
+    /**
+     * Semantic enrichment configuration
+     * Enables MT-friendly PII tags with gender/scope attributes
+     * @default { enabled: false }
+     */
+    semantic?: SemanticConfig;
+    /**
+     * Secrets/credentials detection configuration
+     * Detects API keys, private keys, JWTs, connection strings, etc.
+     * @default { enabled: false }
+     */
+    secrets?: SecretsConfig;
+    /** Key provider for encryption (generates random key if not provided) */
+    keyProvider?: KeyProvider;
+    /**
+     * PII storage provider for automatic session-based persistence
+     * When provided, enables the session() method for automatic PII map storage
+     */
+    piiStorageProvider?: PIIStorageProvider;
+    /** Default policy — merged with SDK defaults, so partial overrides are safe */
+    defaultPolicy?: Partial<AnonymizationPolicy>;
+    /** Model version string */
+    modelVersion?: string;
+    /** Policy version string */
+    policyVersion?: string;
+    /**
+     * Optional callback for validation warnings emitted during anonymization.
+     * When not provided, warnings are silently ignored.
+     * Previously these were logged via console.warn, which pollutes stdout
+     * in TUI-based editors.
+     */
+    onValidationWarning?: (warnings: Array<{
+        code: string;
+        message: string;
+    }>) => void;
+    /**
+     * Tag format configuration for PII placeholder tags.
+     * Controls the delimiters and keyword used in tags.
+     * @default { open: "<", close: "/>", keyword: "PII" } (XML-style)
+     * @example { open: "[[", close: "]]" } for bracket-style tags
+     */
+    tagFormat?: TagFormat;
+}
+/**
+ * Session implementation factory type
+ */
+export type SessionFactory = (anonymizer: Anonymizer, sessionId: string, storage: PIIStorageProvider, keyProvider: KeyProvider) => AnonymizerSession;
+/**
+ * Anonymizer instance
+ * Main class for performing PII anonymization
+ */
+export declare class Anonymizer {
+    private registry;
+    private nerModel;
+    private nerConfig;
+    private semanticConfig;
+    private secretsConfig;
+    private mode;
+    private keyProvider;
+    private piiStorageProvider;
+    private defaultPolicy;
+    private modelVersion;
+    private policyVersion;
+    private initialized;
+    private semanticDataReady;
+    private sessionFactory;
+    private onValidationWarning;
+    private tagFormat;
+    constructor(config?: AnonymizerConfig, sessionFactory?: SessionFactory);
+    /**
+     * Returns the resolved tag format configuration
+     */
+    get resolvedTagFormat(): TagFormat;
+    /**
+     * Initializes the anonymizer
+     * Downloads NER model and semantic data if needed and loads them
+     */
+    initialize(): Promise<void>;
+    /**
+     * Anonymizes text, replacing PII with placeholder tags
+     * @param text - Input text to anonymize
+     * @param locale - Optional locale hint (e.g., 'de-DE', 'en-US')
+     * @param policy - Optional policy override
+     * @param existingPiiMap - Optional existing PII map for session-level ID reuse
+     * @returns Anonymization result with anonymized text and encrypted PII map
+     */
+    anonymize(text: string, locale?: string, policy?: Partial<AnonymizationPolicy>, existingPiiMap?: RawPIIMap): Promise<AnonymizationResult>;
+    /**
+     * Disposes of resources
+     */
+    dispose(): Promise<void>;
+    /**
+     * Gets the recognizer registry
+     */
+    getRegistry(): RecognizerRegistry;
+    /**
+     * Gets the NER model
+     */
+    getNERModel(): INERModel | null;
+    /**
+     * Whether the anonymizer is initialized
+     */
+    get isInitialized(): boolean;
+    /**
+     * Creates a session-bound interface for automatic PII map storage
+     *
+     * @param sessionId - Unique identifier for this session/conversation
+     * @returns AnonymizerSession that auto-saves on anonymize and auto-loads on rehydrate
+     * @throws Error if piiStorageProvider was not configured
+     *
+     * @example
+     * ```typescript
+     * const session = anonymizer.session('chat-123');
+     *
+     * // Anonymize - auto-saves to storage
+     * const result = await session.anonymize('Hello John Smith!');
+     *
+     * // Rehydrate - auto-loads and decrypts
+     * const original = await session.rehydrate(translatedText);
+     * ```
+     */
+    session(sessionId: string): AnonymizerSession;
+}
+/**
+ * Creates an anonymizer with the specified configuration
+ *
+ * @example
+ * ```typescript
+ * // Regex-only (no NER)
+ * const anonymizer = createAnonymizer();
+ *
+ * // With NER (auto-downloads model on first use)
+ * const anonymizer = createAnonymizer({
+ *   ner: { mode: 'quantized' }
+ * });
+ *
+ * // With NER and progress callback
+ * const anonymizer = createAnonymizer({
+ *   ner: {
+ *     mode: 'standard',
+ *     onStatus: (status) => console.log(status),
+ *     onDownloadProgress: (p) => console.log(`${p.file}: ${p.percent}%`)
+ *   }
+ * });
+ * ```
+ */
+export declare function createAnonymizer(config?: AnonymizerConfig, sessionFactory?: SessionFactory): Anonymizer;
+/**
+ * Convenience function for one-off anonymization
+ * Creates a temporary anonymizer with default settings (regex-only)
+ */
+export declare function anonymize(text: string, locale?: string, policy?: Partial<AnonymizationPolicy>): Promise<AnonymizationResult>;
+/**
+ * Quick regex-only anonymization (no NER, faster)
+ */
+export declare function anonymizeRegexOnly(text: string, policy?: Partial<AnonymizationPolicy>): Promise<AnonymizationResult>;
+/**
+ * Full anonymization with NER
+ * Auto-downloads the quantized model on first use
+ *
+ * @example
+ * ```typescript
+ * const result = await anonymizeWithNER(
+ *   'Contact John Smith at john@example.com',
+ *   {
+ *     mode: 'quantized',
+ *     onStatus: console.log
+ *   }
+ * );
+ * ```
+ */
+export declare function anonymizeWithNER(text: string, nerConfig: Omit<NERConfig, "mode"> & {
+    mode?: "standard" | "quantized";
+}, policy?: Partial<AnonymizationPolicy>): Promise<AnonymizationResult>;
+//# sourceMappingURL=anonymizer.d.ts.map

package/dist/core/anonymizer.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"anonymizer.d.ts","sourceRoot":"","sources":["../../src/core/anonymizer.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAEH,OAAO,EACL,mBAAmB,EACnB,mBAAmB,EAEnB,iBAAiB,EAEjB,cAAc,EACd,aAAa,EAEb,OAAO,EACP,SAAS,EAIV,MAAM,mBAAmB,CAAC;AAE3B,OAAO,EAEL,kBAAkB,EAEnB,MAAM,yBAAyB,CAAC;AAGjC,OAAO,EACL,KAAK,SAAS,EAId,KAAK,iBAAiB,EACvB,MAAM,iBAAiB,CAAC;AAEzB,OAAO,EACL,KAAK,YAAY,EAGjB,KAAK,wBAAwB,EAC9B,MAAM,yBAAyB,CAAC;AAIjC,OAAO,EAGL,KAAK,SAAS,EACf,MAAM,uBAAuB,CAAC;AAY/B,OAAO,EAGL,KAAK,WAAW,EACjB,MAAM,oBAAoB,CAAC;AAI5B,YAAY,EAAE,SAAS,EAAE,MAAM,uBAAuB,CAAC;AAGvD,OAAO,KAAK,EACV,kBAAkB,EAClB,iBAAiB,EAClB,MAAM,qBAAqB,CAAC;AAG7B,YAAY,EAAE,kBAAkB,EAAE,iBAAiB,EAAE,CAAC;AAsCtD;;GAEG;AACH,MAAM,WAAW,SAAS;IACxB;;;;;;OAMG;IACH,IAAI,EAAE,YAAY,CAAC;IAEnB;;OAEG;IACH,SAAS,CAAC,EAAE,MAAM,CAAC;IAEnB;;OAEG;IACH,SAAS,CAAC,EAAE,MAAM,CAAC;IAEnB;;;OAGG;IACH,YAAY,CAAC,EAAE,OAAO,CAAC;IAEvB;;OAEG;IACH,kBAAkB,CAAC,EAAE,wBAAwB,CAAC;IAE9C;;OAEG;IACH,QAAQ,CAAC,EAAE,CAAC,MAAM,EAAE,MAAM,KAAK,IAAI,CAAC;IAEpC;;;;OAIG;IACH,UAAU,CAAC,EAAE,OAAO,CAAC,MAAM,CAAC,OAAO,EAAE,MAAM,CAAC,CAAC,CAAC;IAE9C;;;;OAIG;IACH,cAAc,CAAC,EAAE,iBAAiB,CAAC;IAEnC;;;;;OAKG;IACH,OAAO,CAAC,EAAE,OAAO,GAAG,kBAAkB,CAAC;IAEvC;;;OAGG;IACH,kBAAkB,CAAC,EAAE,MAAM,CAAC;IAE5B;;;OAGG;IACH,sBAAsB,CAAC,EAAE,MAAM,CAAC;IAEhC;;;;;;OAMG;IACH,YAAY,CAAC,EAAE,OAAO,CAAC;IAEvB;;;;OAIG;IACH,mBAAmB,CAAC,EAAE,MAAM,CAAC;IAE7B;;;;;OAKG;IACH,mBAAmB,CAAC,EAAE,MAAM,CAAC;IAE7B;;;;;;;OAOG;IACH,kBAAkB,CAAC,EAAE,MAAM,CAAC;IAE5B;;;;;OAKG;IACH,WAAW,CAAC,EAAE,CAAC,IAAI,EAAE;QAAE,cAAc,EAAE,MAAM,CAAC;QAAC,WAAW,EAAE,MAAM,CAAA;KAAE,KAAK,IAAI,CAAC;CAC/E;AAED;;GAEG;AACH,MAAM,WAAW,gBAAgB;IAC/B,yDAAyD;IACzD,QAAQ,CAAC,EAAE,kBAAkB,CAAC;IAE9B;;;;;OAKG;IACH,IAAI,CAAC,EAAE,iBAAiB,CAAC;IAEzB;;;OAGG;IACH,GAAG,CAAC,EAAE,SAAS,CAAC;IAEhB;;;;OAIG;IACH,QAAQ,CAAC,EAAE,cAAc,CAAC;IAE1B;;;;OAIG;IACH,OAAO,CAAC,EAAE,aAAa,CAAC;IAExB,yEAAyE;IACzE,WAAW,CAAC,EAAE,WAAW,CAAC;IAE1B;;;OAGG;IACH,kBAAkB,CAAC,EAAE,kBAAkB,CAAC;IAExC,+EAA+E;IAC/E,aAAa,CAAC,EAAE,OAAO,CAAC,mBAAmB,CAAC,CAAC;IAE7C,2BAA2B;IAC3B,YAAY,CAAC,EAAE,MAAM,CAAC;IAEtB,4BAA4B;IAC5B,aAAa,CAAC,EAAE,MAAM,CAAC;IAEvB;;;;;OAKG;IACH,mBAAmB,CAAC,EAAE,CAAC,QAAQ,EAAE,KAAK,CAAC;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,OAAO,EAAE,MAAM,CAAA;KAAE,CAAC,KAAK,IAAI,CAAC;IAEnF;;;;;OAKG;IACH,SAAS,CAAC,EAAE,SAAS,CAAC;CACvB;AAED;;GAEG;AACH,MAAM,MAAM,cAAc,GAAG,CAC3B,UAAU,EAAE,UAAU,EACtB,SAAS,EAAE,MAAM,EACjB,OAAO,EAAE,kBAAkB,EAC3B,WAAW,EAAE,WAAW,KACrB,iBAAiB,CAAC;AAEvB;;;GAGG;AACH,qBAAa,UAAU;IACrB,OAAO,CAAC,QAAQ,CAAqB;IACrC,OAAO,CAAC,QAAQ,CAA0B;IAC1C,OAAO,CAAC,SAAS,CAAY;IAC7B,OAAO,CAAC,cAAc,CAAiB;IACvC,OAAO,CAAC,aAAa,CAAgB;IACrC,OAAO,CAAC,IAAI,CAAoB;IAChC,OAAO,CAAC,WAAW,CAAqB;IACxC,OAAO,CAAC,kBAAkB,CAA4B;IACtD,OAAO,CAAC,aAAa,CAAsB;IAC3C,OAAO,CAAC,YAAY,CAAS;IAC7B,OAAO,CAAC,aAAa,CAAS;IAC9B,OAAO,CAAC,WAAW,CAAS;IAC5B,OAAO,CAAC,iBAAiB,CAAS;IAClC,OAAO,CAAC,cAAc,CAAwB;IAC9C,OAAO,CAAC,mBAAmB,CAAwE;IACnG,OAAO,CAAC,SAAS,CAAY;gBAEjB,MAAM,GAAE,gBAAqB,EAAE,cAAc,CAAC,EAAE,cAAc;IA4E1E;;OAEG;IACH,IAAI,iBAAiB,IAAI,SAAS,CAEjC;IAED;;;OAGG;IACG,UAAU,IAAI,OAAO,CAAC,IAAI,CAAC;IA6JjC;;;;;;;OAOG;IACG,SAAS,CACb,IAAI,EAAE,MAAM,EACZ,MAAM,CAAC,EAAE,MAAM,EACf,MAAM,CAAC,EAAE,OAAO,CAAC,mBAAmB,CAAC,EACrC,cAAc,CAAC,EAAE,SAAS,GACzB,OAAO,CAAC,mBAAmB,CAAC;IA4I/B;;OAEG;IACG,OAAO,IAAI,OAAO,CAAC,IAAI,CAAC;IAO9B;;OAEG;IACH,WAAW,IAAI,kBAAkB;IAIjC;;OAEG;IACH,WAAW,IAAI,SAAS,GAAG,IAAI;IAI/B;;OAEG;IACH,IAAI,aAAa,IAAI,OAAO,CAE3B;IAED;;;;;;;;;;;;;;;;;OAiBG;IACH,OAAO,CAAC,SAAS,EAAE,MAAM,GAAG,iBAAiB;CAmD9C;AAED;;;;;;;;;;;;;;;;;;;;;;GAsBG;AACH,wBAAgB,gBAAgB,CAC9B,MAAM,CAAC,EAAE,gBAAgB,EACzB,cAAc,CAAC,EAAE,cAAc,GAC9B,UAAU,CAEZ;AAED;;;GAGG;AACH,wBAAsB,SAAS,CAC7B,IAAI,EAAE,MAAM,EACZ,MAAM,CAAC,EAAE,MAAM,EACf,MAAM,CAAC,EAAE,OAAO,CAAC,mBAAmB,CAAC,GACpC,OAAO,CAAC,mBAAmB,CAAC,CAS9B;AAED;;GAEG;AACH,wBAAsB,kBAAkB,CACtC,IAAI,EAAE,MAAM,EACZ,MAAM,CAAC,EAAE,OAAO,CAAC,mBAAmB,CAAC,GACpC,OAAO,CAAC,mBAAmB,CAAC,CAQ9B;AAED;;;;;;;;;;;;;;GAcG;AACH,wBAAsB,gBAAgB,CACpC,IAAI,EAAE,MAAM,EACZ,SAAS,EAAE,IAAI,CAAC,SAAS,EAAE,MAAM,CAAC,GAAG;IAAE,IAAI,CAAC,EAAE,UAAU,GAAG,WAAW,CAAA;CAAE,EACxE,MAAM,CAAC,EAAE,OAAO,CAAC,mBAAmB,CAAC,GACpC,OAAO,CAAC,mBAAmB,CAAC,CAe9B"}