@framers/agentos 0.1.149 → 0.1.151

package/README.md

@@ -6,7 +6,7 @@
 
 # AgentOS
 
-**TypeScript runtime for autonomous AI agents — multimodal RAG, cognitive memory, streaming guardrails, voice pipeline, and emergent multi-agent orchestration.**
+**Build autonomous AI agents with adaptive intelligence and emergent behaviors. Multimodal RAG, cognitive memory, streaming guardrails, voice pipeline, and 21 LLM providers. Open-source TypeScript runtime.**
 
 [![npm version](https://img.shields.io/npm/v/@framers/agentos?style=flat-square&logo=npm&color=cb3837)](https://www.npmjs.com/package/@framers/agentos)
 [![CI](https://img.shields.io/github/actions/workflow/status/framersai/agentos/ci.yml?branch=master&style=flat-square&logo=github&label=CI)](https://github.com/framersai/agentos/actions/workflows/ci.yml)
@@ -24,6 +24,8 @@
 ## Table of Contents
 
 - [Overview](#overview)
+- [Adaptive Intelligence](#adaptive-intelligence)
+- [Emergent Behaviors](#emergent-behaviors)
 - [Quick Start](#quick-start)
   - [1. Generate Text (Provider-First)](#1-generate-text-provider-first)
   - [2. Agent with Personality & Memory](#2-agent-with-personality--memory)
@@ -91,7 +93,9 @@
 
 ## Overview
 
-`@framers/agentos` is an open-source TypeScript AI agent runtime for building, deploying, and managing production AI agents. It provides multimodal RAG with cognitive memory (Ebbinghaus decay, 8 neuroscience-backed mechanisms including reconsolidation, retrieval-induced forgetting, involuntary recall, metacognitive FOK, temporal gist extraction, schema encoding, source confidence decay, and emotion regulation — all HEXACO personality-modulated), multi-agent orchestration, 37 channel adapters, 6 guardrail packs (PII redaction, ML classifiers, topicality, code safety, grounding guard, content policy rewriter) with prompt injection defense, 21 LLM providers, and 72 curated skills. Self-hostable and production-ready, it handles the full lifecycle from prompt construction through tool execution, safety evaluation, and streaming response delivery.
+`@framers/agentos` is an open-source TypeScript AI agent runtime for building autonomous AI agents with **adaptive intelligence** and **emergent behaviors**. Agents continuously improve through meta-reflective prompt adaptation, self-evaluating response quality, personality-modulated cognition, and autonomous memory consolidation. At runtime, agents forge new tools, adapt their personality, compose workflows, and promote reliable capabilities for cross-agent reuse.
+
+The runtime provides multimodal RAG with cognitive memory (Ebbinghaus decay, 8 neuroscience-backed mechanisms including reconsolidation, retrieval-induced forgetting, involuntary recall, metacognitive FOK, temporal gist extraction, schema encoding, source confidence decay, and emotion regulation ��� all HEXACO personality-modulated), multi-agent orchestration, 37 channel adapters, 6 guardrail packs (PII redaction, ML classifiers, topicality, code safety, grounding guard, content policy rewriter) with prompt injection defense, 21 LLM providers, and 72 curated skills. Self-hostable and production-ready, it handles the full lifecycle from prompt construction through tool execution, safety evaluation, and streaming response delivery.
 
 **Key facts:**
 
@@ -154,6 +158,34 @@
 
 ---
 
+## Adaptive Intelligence
+
+AgentOS agents continuously improve their behavior without retraining through five interconnected mechanisms:
+
+| Mechanism | How It Works |
+|-----------|-------------|
+| **Meta-reflective prompt adaptation** | The PromptBuilder assembles a different system prompt every turn, dynamically incorporating personality traits, mood state, conversation history, retrieved memories, and available tools. No two turns use the same prompt. |
+| **Self-evaluating response quality** | The `self_evaluate` tool scores the agent's own output and adjusts parameters (temperature, verbosity, personality expression) in real time, creating a continuous feedback loop. |
+| **Personality-modulated cognition** | HEXACO traits modulate how the agent processes information. High openness increases creative associations during memory retrieval; high conscientiousness strengthens retrieval-induced forgetting of irrelevant data. |
+| **Autonomous memory consolidation** | The `ConsolidationLoop` prunes weak memories, strengthens frequently-accessed ones, and derives new insights from memory clusters. The agent's knowledge base improves over time without explicit training. |
+| **QueryRouter tiered classification** | The system adapts retrieval depth based on query complexity. Simple questions get fast keyword lookup; complex questions trigger full hybrid RAG with deep research. |
+
+---
+
+## Emergent Behaviors
+
+AgentOS agents develop new capabilities at runtime rather than having them explicitly programmed:
+
+| Capability | How It Works |
+|-----------|-------------|
+| **Runtime tool forging** | Agents create new tools on the fly via `forge_tool`. The `EmergentCapabilityEngine` uses sandboxed JavaScript execution and LLM-as-judge evaluation to safely create, test, and promote tools. |
+| **Self-improving personality** | Agents adapt their HEXACO personality traits within bounded limits via `adapt_personality`. Mutations persist with Ebbinghaus decay — strong repeated adaptations stick, while weak ones fade naturally. |
+| **Dynamic skill management** | Agents enable or disable skills at runtime via `manage_skills`, adapting their behavioral repertoire to the task at hand. |
+| **Composable workflow creation** | Agents compose registered tools into multi-step pipelines via `create_workflow`, building new capabilities from existing building blocks. |
+| **Tiered tool promotion** | Forged tools progress through session, agent, and shared tiers. Tools that prove reliable (5+ successful uses, >0.8 confidence score) auto-promote for cross-agent reuse. |
+
+---
+
 ## Quick Start
 
 ```bash

package/knowledge/platform-corpus.json

@@ -179,6 +179,36 @@
     "content": "AgentOS implements 8 research-backed memory mechanisms: reconsolidation, retrieval-induced forgetting (RIF), involuntary recall, feeling-of-knowing (FOK), temporal gist extraction, schema encoding, source confidence decay, and emotion regulation. All are HEXACO personality-modulated.",
     "category": "faq"
   },
+  {
+    "id": "faq:setup-gmail",
+    "heading": "How to set up Gmail integration",
+    "content": "To set up Gmail: 1) User needs a Google Cloud project with Gmail API enabled. 2) Create OAuth 2.0 credentials (Desktop app type). 3) Google downloads a client_secret_*.json file. 4) The file contains {\"installed\": {\"client_id\": \"...\", \"client_secret\": \"...\"}}. 5) Extract client_id and client_secret. 6) Set as GOOGLE_CLIENT_ID and GOOGLE_CLIENT_SECRET env vars, or pass to wunderland connect gmail --credentials <path>. 7) Run wunderland connect gmail to complete OAuth flow. If the user says they downloaded a credentials file, use shell_execute to find it: ls -t ~/Downloads/client_secret*.json | head -1. Then use file_read to parse it and extract the values.",
+    "category": "faq"
+  },
+  {
+    "id": "faq:setup-credentials-general",
+    "heading": "How to help users set up API credentials",
+    "content": "When a user needs to set up credentials for any service: 1) Check what the extension needs using discover_capabilities or the platform knowledge. 2) Each extension lists requiredSecrets in the capability catalog. 3) Common patterns: API keys go in env vars or agent.config.json secrets. OAuth flows (Gmail, Slack, Discord) use wunderland connect <service>. 4) If the user has downloaded a credentials file, offer to read it: use shell_execute to find files (ls ~/Downloads/*.json) and file_read to parse them. 5) For Google services, the file is client_secret_*.json with format {\"installed\": {\"client_id\": \"...\", \"client_secret\": \"...\"}}. 6) For other services, API keys are usually plain text strings from their dashboard. 7) After extracting credentials, either set env vars with shell_execute (export KEY=value >> ~/.bashrc) or write to ~/.wunderland/config.json.",
+    "category": "faq"
+  },
+  {
+    "id": "faq:find-credential-files",
+    "heading": "How to find credential files on the user system",
+    "content": "When a user says they downloaded credentials or a secret file: 1) Check common download locations: ~/Downloads/, ~/Desktop/, current directory. 2) Use shell_execute with ls -t to find recent files: ls -t ~/Downloads/*.json ~/Downloads/*.txt 2>/dev/null | head -10. 3) For Google: ls -t ~/Downloads/client_secret*.json | head -1. 4) For AWS: look for ~/.aws/credentials or downloaded CSV from IAM console. 5) For SSH keys: look for ~/.ssh/id_*.pub. 6) Always confirm with the user before reading any file. 7) Parse the file with file_read and extract the relevant values. 8) Never store raw credential files — extract the needed values and save them securely.",
+    "category": "faq"
+  },
+  {
+    "id": "faq:extension-credentials",
+    "heading": "What credentials each extension needs",
+    "content": "Common extension credential requirements: Gmail: GOOGLE_CLIENT_ID + GOOGLE_CLIENT_SECRET + GOOGLE_REFRESH_TOKEN (OAuth). GitHub: GITHUB_TOKEN (Personal Access Token from github.com/settings/tokens). Slack: SLACK_BOT_TOKEN + SLACK_SIGNING_SECRET. Discord: DISCORD_BOT_TOKEN. Telegram: TELEGRAM_BOT_TOKEN (from @BotFather). OpenAI: OPENAI_API_KEY (from platform.openai.com/api-keys). Anthropic: ANTHROPIC_API_KEY. Serper: SERPER_API_KEY (from serper.dev). ElevenLabs: ELEVENLABS_API_KEY. Stability: STABILITY_API_KEY. Replicate: REPLICATE_API_TOKEN. Runway: RUNWAY_API_KEY. Suno: SUNO_API_KEY. Brave: BRAVE_API_KEY. NewsAPI: NEWSAPI_API_KEY. Giphy: GIPHY_API_KEY.",
+    "category": "faq"
+  },
+  {
+    "id": "faq:wunderland-connect",
+    "heading": "wunderland connect command reference",
+    "content": "wunderland connect <service> starts an OAuth flow for services that need it. Supported: gmail (Google OAuth to Gmail API access), whatsapp (Twilio or Meta Cloud API), slack (OAuth via Rabbithole), signal (signal-cli setup wizard). For Gmail: wunderland connect gmail [--credentials path/to/client_secret.json]. The --credentials flag reads a Google Cloud Console JSON file and extracts client_id + client_secret automatically. Without it, needs GOOGLE_CLIENT_ID and GOOGLE_CLIENT_SECRET env vars. After OAuth completes, tokens are saved to ~/.wunderland/config.json and automatically loaded on next wunderland chat/start.",
+    "category": "faq"
+  },
   {
     "id": "api:generateText",
     "heading": "generateText() API",

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@framers/agentos",
-  "version": "0.1.149",
+  "version": "0.1.151",
   "description": "Modular AgentOS orchestration library",
   "type": "module",
   "main": "dist/index.js",