Observed Signal · Apr 7, 2026 · Technical Release · Source: DEV Community · Impact: 2/5 · Sentiment: Positive
selectools: Lightweight Open-Source AI Agent Framework
selectools is an open-source Python library for building production-ready AI agents that emphasizes simple function-based tool calling, local traceability without a paid SaaS, built-in guardrails, plain-Python multi-agent routing, and one-command deployment. The project includes a browser-based visual agent builder hosted on GitHub Pages, extensive tests (4,612 tests at 95% coverage), security hardening from a pre-launch audit, dozens of runnable examples and evaluators, and an Apache-2.0 license. The author positions selectools as a smaller, developer-friendly alternative to larger frameworks (e.g., LangChain) for teams that prioritize direct control, observability, and lightweight orchestration across OpenAI, Anthropic, Gemini and Ollama integrations.
Introduces a lightweight, open-source agent framework with production-focused features (local tracing, guardrails, multi-agent orchestration and a browser visual builder) that can lower development friction for teams building LLM agents; however it is a smaller project with fewer integrations than larger frameworks, limiting near-term industry impact.
Track Anthropic Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- selectools is an open-source Python library for building AI agents and is distributed under the Apache-2.0 license.
- The library supports tool calling across OpenAI, Anthropic, Gemini and Ollama and favors plain-Python routing and function-based pipelines (e.g., AgentGraph.chain and use of | operator pipelines).
- A browser-based visual agent builder is deployed at https://selectools.dev/builder/ (hosted on GitHub Pages) and can export agent definitions to YAML or Python.
- Project metrics: 4,612 tests at 95% coverage; 9 critical security bugs fixed in a pre-launch audit; 76 runnable examples; 50 built-in evaluators; 152 model definitions with pricing data.
- One-command deployment is provided (selectools serve agent.yaml) offering HTTP endpoints, SSE streaming, and a chat playground without requiring separate FastAPI setup.
Connected Companies & Entities
6 Entities mappedOntology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
SoloEngine: Low-code Platform for Autonomous AI Loops
SoloEngine is an open-source, low-code Agentic AI development platform that packages Loop Engineering primitives into visual modules. Published June 26, 2026, the project offers a browser canvas to drag Agent nodes, one-click compilation into an executable Agent DAG, and an Auto-run mode where Agents execute unified ReAct loops (Think → Act → Observe → Repeat). SoloEngine emphasizes progressive disclosure to reduce token consumption (claimed >85% savings), MCP (Model Context Protocol) tool integration for connecting real business systems, and a multi-model adapter layer for OpenAI, Anthropic, Ollama, DeepSeek, Qwen and ChatGLM. The repo is available on GitHub under Apache 2.0; a v0.4 milestone will add one-click packaging to publish Agent teams as deployable products.
Open Engine: AI Agent Handoffs Without Humans
The author announces Open Engine, a practical framework and set of copy-paste templates to let AI agents hand off work across models and tools without requiring a human to carry the state. The project focuses on the integration layer — preserving sources, limits, and provenance as a task moves between agents (Claude, Codex, ChatGPT, browser agents) and collaboration tools (Slack, Linear, calendar). Open Engine includes a shared task list, a seven-part task record, a compact accountability “receipt,” and a nine-question one-loop audit designed to let an agent claim, pause, resume, and finish tasks with evidence. The release aims to solve the operational friction of multi-model, multi-tool workflows rather than kingmaking among models, and positions Open Engine alongside other orchestration projects such as OpenClaw, Hermes, and Symphony.
Open Executive: Open-source AI Aiming to Act as CEO
Open Executive is an open-source AI tool from Sente Labs AI, published on GitHub, that simulates a company's entire executive team. A primary 'CEO' agent is supported by sub-agents such as CFO, CHRO, and legal advisor, all with configurable permissions, budgets, and approval requirements. The system uses Anthropic Claude models (Sonnet 4.6 as standard, Opus 4.7 for reasoning) via the Anthropic API, which incurs costs. It integrates through a web interface and chat platforms (Slack, email, Telegram, Google Chat, Discord). Decisions are stored in long-term memory, allowing the tool to retain rationale and draw on past decisions, while users can inspect each model's reasoning. The system can autonomously make decisions, as demonstrated in an electric vehicle company scenario where it chose to maintain prices despite a competitor's cut.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.