# DSPy Micro Agent — CLI + FastAPI + Evals

> evalops/dspy-micro-agent is a minimal agent runtime (CLI + FastAPI + eval harness); verified 73★ and documents `micro-agent ask` and `run_evals.py --n 50`.

## Install

Copy the content below into your project:

## Quick Use

```bash
uv venv && source .venv/bin/activate
uv pip install -e .
cp .env.example .env  # set OPENAI_API_KEY or configure Ollama
micro-agent ask --question "What's 2*(3+5)?" --utc
python evals/run_evals.py --n 50
```

## Intro

evalops/dspy-micro-agent is a minimal agent runtime (CLI + FastAPI + eval harness); verified 73★ and documents `micro-agent ask` and `run_evals.py --n 50`.

**Best for:** Builders who want a small, readable Plan/Act loop with traces, plus quick evals before adding complexity

**Works with:** Python 3.10+, OpenAI or Ollama providers, CLI + FastAPI deployments

**Setup time:** 10-25 minutes

### Key facts (verified)

- GitHub: 73 stars · 6 forks · pushed 2026-04-25.
- License: MIT · owner avatar + repo URL verified via GitHub API.
- README-backed entrypoint: `micro-agent ask --question "What's 2*(3+5)?" --utc`.

## Main

- Treat traces as a first-class artifact: README stores JSONL traces under `traces/` and includes replay support.

- Use the eval harness early (`--n 50`) to guard against regressions when you add tools, prompts, or provider switches.

- Start with the CLI, then graduate to the HTTP API when you need multi-client access or UI integrations.

### Source-backed notes

- README quickstart shows `micro-agent ask ... --utc`, a FastAPI server via `uvicorn`, and evals via `python evals/run_evals.py --n 50`.
- Docs describe provider config for OpenAI and Ollama, plus tracing under `traces/<id>.jsonl`.
- README lists eval metrics like success_rate/avg_latency_sec/avg_cost_usd and notes usage/cost capture can be best-effort.

### FAQ

- **Is this a full framework?**: No — README frames it as a minimal runtime + DSPy modules you can read end-to-end.
- **Can I run it without OpenAI?**: Yes — README includes an Ollama provider path via env vars.
- **How do I keep changes safe?**: Use the built-in eval harness and store/replay traces to compare behavior over time.

## Source & Thanks

> Source: https://github.com/evalops/dspy-micro-agent
> License: MIT
> GitHub stars: 73 · forks: 6

---

<!-- ZH -->

## Quick Use

```bash
uv venv && source .venv/bin/activate
uv pip install -e .
cp .env.example .env  # set OPENAI_API_KEY or configure Ollama
micro-agent ask --question "What's 2*(3+5)?" --utc
python evals/run_evals.py --n 50
```

## Intro

evalops/dspy-micro-agent 提供最小 agent runtime：CLI + FastAPI 服务 + 小型 eval harness；已验证 73★，README 展示 `micro-agent ask` 与 `run_evals.py --n 50`。

**Best for:** 想先用小而清晰的 Plan/Act 循环 + trace，再逐步扩展复杂度的开发者

**Works with:** Python 3.10+，支持 OpenAI/Ollama provider，适配 CLI 与 FastAPI 部署

**Setup time:** 10-25 minutes

### Key facts (verified)

- GitHub：73 stars · 6 forks；最近更新 2026-04-25。
- 许可证：MIT；作者头像与仓库链接均已通过 GitHub API 复核。
- README 中可对照的入口命令：`micro-agent ask --question "What's 2*(3+5)?" --utc`。

## Main

- 把 trace 当成一等产物：README 将 JSONL trace 存在 `traces/`，并提供回放能力。

- 尽早用 eval harness（`--n 50`）做回归门禁，避免加工具/改 prompt/换 provider 后变差。

- 先用 CLI 跑通；需要多端接入或 UI 集成时再上 HTTP API。

### Source-backed notes

- README quickstart 包含 `micro-agent ask ... --utc`、`uvicorn` 启动 FastAPI、以及 `python evals/run_evals.py --n 50`。
- 文档说明 OpenAI/Ollama provider 配置，并将 trace 写入 `traces/<id>.jsonl`。
- README 列出 eval 指标（success_rate、avg_latency_sec、avg_cost_usd 等），并说明成本统计可能是 best-effort。

### FAQ

- **它是完整框架吗？**：不是。README 强调它是最小 runtime + DSPy 模块，便于端到端阅读。
- **不接 OpenAI 能跑吗？**：能。README 提供 Ollama 的环境变量配置路径。
- **如何让改动更安全？**：用自带 eval harness 做门禁，并保存/回放 traces 对比行为变化。

## Source & Thanks

> Source: https://github.com/evalops/dspy-micro-agent
> License: MIT
> GitHub stars: 73 · forks: 6


---
Source: https://tokrepo.com/en/workflows/dspy-micro-agent-cli-fastapi-evals
Author: Agent Toolkit