camelagi
Personal AI agent powered by Claude Agent SDK โ manage everything from Telegram
Why this rank:Healthy release cadenceRelease freshness
Description
Personal AI agent powered by Claude Agent SDK โ manage everything from Telegram
Release History
| Version | Changes | Urgency | Date |
|---|---|---|---|
| main@2026-05-16 | Latest activity on main branch | Low | 5/16/2026 |
| 0.5.41 | Imported from npm (0.5.41) | Low | 4/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.16 | Debug build | Low | 3/21/2026 |
| v0.5.15 | Debug build with detailed SDK error logging | Low | 3/21/2026 |
| v0.5.14 | Debug build with try-catch around query() and stderr logging for SDK path + errors | Low | 3/21/2026 |
| v0.5.13 | ## Fix - **Eliminated all spread operators** in the agent execution path (orchestrate.ts + agent-sdk.ts) โ replaced with step-by-step object construction - **Added `[DEBUG] Full error:` logging** โ if the spread error persists, the server console will now print the full error object and stack trace | Low | 3/21/2026 |
| v0.5.12 | Debug build โ logs full stack trace on agent errors to pinpoint spread crash | Low | 3/21/2026 |
| v0.5.11 | ## Fix - **Guard all config property spreads** โ configs created before certain features (compaction, tools, hooks, approvals, retry, MCP) existed could have undefined values that crash when spread. All now safely default to `{}` or hardcoded fallbacks. | Low | 3/21/2026 |
| v0.5.10 | ## Fix - **Fix "Spread syntax requires ...iterable" crash** โ `config.mcp.servers` could be undefined for configs created before MCP support was added. Now safely defaults to `{}`. | Low | 3/21/2026 |
| v0.5.9 | ## Improvement - **Pairing notifications in server console** โ when someone sends a message to your bot and needs approval, the server now prints: ``` Pairing request from @username Run: camel pairing ``` So you know immediately when to open another terminal and approve. | Low | 3/21/2026 |
| v0.5.8 | ## Fix - **Lazy-load agent SDK** โ `camel serve`, `camel setup`, and admin bot no longer crash with "Cannot find module @anthropic-ai/claude-agent-sdk". The SDK (which has native binaries) is now loaded only when a message is actually processed, not at startup. | Low | 3/21/2026 |
| v0.5.7 | ## Fixes - **Version sync** โ binary now correctly reports its version (was stuck at 0.5.4) - **Setup crash fix** โ empty bot token no longer crashes - **Token retry** โ invalid bot token prompts "Try again?" instead of silently skipping - **Setup UX** โ shows checkmarks for what's already configured (API, Telegram) - **No more agent SDK crash** โ setup starts only the admin bot for pairing, not the full server - **Install script** โ sudo errors no longer hidden, tells user to restart terminal | Low | 3/21/2026 |
| v0.5.6 | ## Fixes - **Version sync** โ `version.ts` now matches `package.json` (was stuck at 0.5.4) - **Setup crash fix** โ empty bot token no longer crashes with `undefined.trim()` - **Token retry** โ invalid bot token now prompts "Try again?" instead of silently skipping - **Setup UX** โ shows checkmarks for what's already configured (API, Telegram) and what's left - **Install script** โ sudo errors no longer hidden, tells user to restart terminal if needed | Low | 3/21/2026 |
| v0.5.5 | ## What's new - **README rewrite** โ accurate features, all command tables, architecture section - **Install script fix** โ sudo errors no longer swallowed, tells user to restart terminal when needed - **Setup fix** โ invalid bot token now retries instead of silently skipping, fixed crash on empty token - **License** โ copyright updated to Nawaf Almutairi & Brickell Lab, Kuwait | Low | 3/21/2026 |
| v0.5.4 | ## What's New - **`/agent <id>`** โ View and edit agent config (model, thinking, effort, max turns) directly from Telegram with inline buttons - **Agent cloning** โ Clone button copies config + workspace files (SOUL.md, TOOLS.md, MEMORY.md, memory/) - **Error alerts** โ Agent errors forwarded to admin bot (throttled to 1 per agent per 5 min) - **`/usage`** โ Per-agent token usage + estimated cost summary (30+ models in pricing table) - **`camel status`** โ CLI command for system health overview | Low | 3/20/2026 |
| v0.5.3 | ## What's New ### Full MCP Support All three transport types from the Claude Agent SDK are now supported: - **HTTP** โ Remote MCP servers via URL (e.g. Supabase, Claude Docs) - **SSE** โ Streaming MCP servers via URL - **stdio** โ Local command-based MCP servers (e.g. GitHub, Postgres) ### Per-Agent MCP Servers Each agent can have its own MCP servers, merged with global servers at runtime. ### Telegram `/mcp` Wizard Available in both admin and agent bots: - **Add** โ Guided wizard with type | Low | 3/19/2026 |
| v0.5.2 | ## What's New ### Fix Agent SDK for non-Claude models via OpenRouter Non-Claude models (e.g. `x-ai/grok-4-fast`) now work correctly through the Claude Agent SDK path when routed via OpenRouter. Previously, using any non-Claude model would fail with "Claude Code executable not found" or return empty responses. **Fixes:** - Pass full `process.env` to SDK subprocess so it can find `node` via `PATH` - Handle `assistant` message type emitted by non-Claude models (instead of `result`) - Gracefully r | Low | 3/19/2026 |
| v0.5.1 | ## v0.5.1 - Unified `camel setup` โ replaces bootstrap, asks TUI/Telegram/Both - Channel system: Telegram + Discord with shared handler - MCP server support - Telegram: markdown rendering, inline buttons for /think /effort, voice transcription - Progressive skill loading - Session export, shared sessions across channels - macOS app: install detection, todo card, no embedded Node.js - Standalone binary install via curl (no Node.js required) - OTP removed โ approve = instant access | Low | 3/18/2026 |
| v0.5.0 | ## CamelAGI v0.5.0 Personal AI assistant powered by Claude Agent SDK. ### Install ```bash curl -fsSL https://raw.githubusercontent.com/inawafalm/CamelAGI/main/install.sh | bash ``` Or via npm: `npm install -g camelagi` | Low | 3/18/2026 |
Dependencies & License Audit
Loading dependencies...
Similar Packages
alvin-botAlvin Bot โ Your personal AI agent on Telegram, WhatsApp, Discord, Signal, and Web.v4.25.1
@mediconsol/agent-core๋ฉ๋์ฝ์ AI Agent ์ฝ์ด ๋ชจ๋ โ ๋ชจ๋ ๋ฉ๋์ฝ์ ํ๋ก์ ํธ์ ํ์ฌ ๊ฐ๋ฅํ ์์ด์ ํธ ์ธํ๋ผ0.1.6
fn-aiAI agent toolkit with pluggable provider, tool, and session store layers. ESM, TypeScript-first.0.6.0
krisspy-aiUnified AI Agent Library - Supports Claude Agent SDK, Codex SDK, and proxy providers (OpenAI, Gemini, Z.AI)1.1.67
@lucashca/claudecontrolAI Agent Dashboard โ manage multiple Claude Code agents across projects0.5.31
More in AI Agents
claude-memA Claude Code plugin that automatically captures everything Claude does during your coding sessions, compresses it with AI (using Claude's agent-sdk), and injects relevant context back into future ses
AIGC-Interview-Bookใไธๅนด้ข่ฏไบๅนดๆจกๆใAIGC็ฎๆณๅทฅ็จๅธ้ข่ฏ็ง็ฑใๆถต็AIGCใLLMๅคงๆจกๅใAI Agentใไผ ็ปๆทฑๅบฆๅญฆไน ใ่ชๅจ้ฉพ้ฉถใๆบๅจๅญฆไน ใ่ฎก็ฎๆบ่ง่งใ่ช็ถ่ฏญ่จๅค็ใๅผบๅๅญฆไน ใๅคงๆฐๆฎๆๆใๅ
ท่บซๆบ่ฝใๅ
ๅฎๅฎใAGI็ญAI่กไธ้ข่ฏ็ฌ่ฏๅนฒ่ดง็ป้ชไธๆ ธๅฟ็ฅ่ฏใ
letta-codeThe memory-first coding agent
nano-banana-pro-prompts-recommend-skillAI skill for OpenClaw & Claude Code โ recommend from 10000+ Nano Banana Pro (Gemini) image prompts. Smart search by use case, content remix, sample images.