Flue: A TypeScript Framework for Building Autonomous Coding Agents

Flue is a TypeScript framework for building autonomous agents using a programmable harness architecture — the same pattern behind Claude Code and Codex. It lets you define agents that plan, gather context, write files, spawn subagents, and run shell commands, all within a sandboxed environment.
Key Features
- Agent Harness: Combines a model (e.g., Anthropic Claude Sonnet 4-6) with a harness for tools like filesystem read/write, grep, glob, bash execution, and network access.
- Sessions: Persistent contexts for tracking work, similar to Claude Code or Codex sessions.
- Skills: Reusable workflows with structured output using Valibot for runtime validation.
- Sandbox: Built-in zero-config virtual sandbox, or connect your own remote sandbox. Fine-grained control over environment variables and token exposure.
- Deployment: Agents bundle into an HTTP server for remote use, or run via CLI (
flue run) for local tasks and CI.
Example: AI Issue Triage in 22 Lines
import type { FlueContext } from '@flue/sdk/client';
import { Octokit } from '@octokit/core';
import * as v from 'valibot';
export default async function ({ init, payload, env }: FlueContext) {
const { issueNumber } = payload;
const agent = await init({ model: 'anthropic/claude-opus-4-7' });
const session = await agent.session();
const triage = await session.skill('triage', {
args: { issueNumber },
result: v.object({
severity: v.picklist(['low', 'medium', 'high', 'critical']),
reproducible: v.boolean(),
summary: v.string(),
}),
});
const body = **Severity:** ${triage.severity}\n**Reproducible:** ${triage.reproducible}\n\n${triage.summary};
await (new Octokit({ auth: env.GITHUB_TOKEN })).request(
'POST /repos/{owner}/{repo}/issues/{num}/comments',
{ owner: 'withastro', repo: 'flue', num: issueNumber, body },
);
}
Who It's For
Developers building custom AI agents for code review, issue triage, data analysis, customer support, or coding automation who want full control over the agent stack without renting a third-party tool.
📖 Read the full source: HN LLM Tools
👀 See Also

Code Evolution Method Triples LLM Performance on ARC-AGI-2 Benchmark
Researchers achieved a 2.8x improvement on the ARC-AGI-2 benchmark using code evolution with open-weight models, reaching 34% accuracy at $2.67 per task. The same method pushed Gemini 3.1 Pro to 95% accuracy at $8.71 per task.

Aired: A Claude Code Skill for Instant HTML Publishing to Live URLs
Aired is an open-source tool that publishes HTML to a live URL in 2 seconds via Claude Code skills or MCP servers. It requires no signup, deployment configuration, or installation for web-based AI tools, and works with Claude Code, Cursor, VS Code, Codex, and Windsurf.

CAP: Claude Code Statusline Plugin That Installs with /plugin install
CAP (Claude Allowance Pulse) is a statusline plugin for Claude Code that installs via /plugin install without npm, curl, or jq. It displays model usage, session and weekly limits, context window usage, and session cost in the terminal.

Bifrost AI Gateway: Open-Source Tool Addresses AI Infrastructure Gaps
Bifrost is an open-source Go-based LLM gateway that provides automatic failover between providers, budget caps that reject requests, audit logging, and hooks for evaluation. Benchmarks show it's ~50x faster than LiteLLM at high throughput.