Agent-Field/SWE-AF Instruction Manual · Repository Routing Wiki

What Agent-Field/SWE-AF is—in plain language

Public-beta autonomous engineering-team runtime coordinating planning, coding, review, QA, merge, verification, and replanning agents.

Mental model: This repository behaves like a service or multi-process application. One command starts it, a browser/API/agent connects to it, and the Terminal process must remain running until you stop it.

InterfacesCLI, HTTP API

Primary languagePython

Typical useRun high-autonomy single/multi-repository engineering programs with parallel worktrees, checkpoints, retries, and model-per-role selection.

Your first 15 minutes

Complete the linked Installation Guide and its verification step.
Create a disposable test folder or use non-sensitive sample data.
Run the small example below and observe what files, ports, or prompts appear.
Read the result before approving writes, network calls, account access, or costs.
Only then repeat the workflow with a real project.

Starter workflow

af call swe-planner.build --in '{"goal": "Add a small documented feature", "repo_url": "https://github.com/OWNER/REPOSITORY"}'

Open Installation Guide Open upstream README

Complete indexed functionality map

Each card below corresponds to a component identified in the repository research. Open a component record for its path, purpose, capabilities, dependencies, risks, and relationships.

swe-planner.build

service

Top-level autonomous build entry.

README.md

Open component record →

Planning stack

agent

PM/architecture/lead planning.

README.md

Open component record →

Coder/reviewer/QA

agent

Implement and validate.

README.md

Open component record →

Merger/verifier

agent

Integrate and verify acceptance.

README.md

Open component record →

Advisor/replanner

agent

Adapt after failures.

README.md

Open component record →

Upstream feature-by-feature guide

These capabilities are derived from the current first-party README and supporting documentation. Names follow upstream terminology so you can search the source documentation precisely.

Upstream capability

Autonomous Engineering Team Runtime Built on AgentField

Upstream treats this as a distinct part of Agent-Field/SWE-AF. In plain language, use this area when your task concerns autonomous engineering team runtime built on agentfield. Begin with the documented default, test it on disposable input, and open the source section for its current options and limitations.

Upstream capability

One-Call DX

Upstream treats this as a distinct part of Agent-Field/SWE-AF. In plain language, use this area when your task concerns one-call dx. Begin with the documented default, test it on disposable input, and open the source section for its current options and limitations.

Upstream capability

Operating Modes

Upstream treats this as a distinct part of Agent-Field/SWE-AF. In plain language, use this area when your task concerns operating modes. Begin with the documented default, test it on disposable input, and open the source section for its current options and limitations.

Upstream capability

Single-Repository Mode

Upstream treats this as a distinct part of Agent-Field/SWE-AF. In plain language, use this area when your task concerns single-repository mode. Begin with the documented default, test it on disposable input, and open the source section for its current options and limitations.

Upstream capability

Multi-Repository Mode

Upstream treats this as a distinct part of Agent-Field/SWE-AF. In plain language, use this area when your task concerns multi-repository mode. Begin with the documented default, test it on disposable input, and open the source section for its current options and limitations.

Upstream capability

Autonomous Build Spotlight

Upstream treats this as a distinct part of Agent-Field/SWE-AF. In plain language, use this area when your task concerns autonomous build spotlight. Begin with the documented default, test it on disposable input, and open the source section for its current options and limitations.

Upstream capability

Why SWE-AF

Upstream treats this as a distinct part of Agent-Field/SWE-AF. In plain language, use this area when your task concerns why swe-af. Begin with the documented default, test it on disposable input, and open the source section for its current options and limitations.

Upstream capability

In Action

Upstream treats this as a distinct part of Agent-Field/SWE-AF. In plain language, use this area when your task concerns in action. Begin with the documented default, test it on disposable input, and open the source section for its current options and limitations.

Upstream capability

Adaptive Factory Control

Upstream treats this as a distinct part of Agent-Field/SWE-AF. In plain language, use this area when your task concerns adaptive factory control. Begin with the documented default, test it on disposable input, and open the source section for its current options and limitations.

Upstream capability

Deploy with Railway (fastest)

Upstream treats this as a distinct part of Agent-Field/SWE-AF. In plain language, use this area when your task concerns deploy with railway (fastest). Begin with the documented default, test it on disposable input, and open the source section for its current options and limitations.

Upstream capability

1. Requirements (local)

Upstream treats this as a distinct part of Agent-Field/SWE-AF. In plain language, use this area when your task concerns 1. requirements (local). Begin with the documented default, test it on disposable input, and open the source section for its current options and limitations.

Upstream capability

4. Trigger a build

Upstream treats this as a distinct part of Agent-Field/SWE-AF. In plain language, use this area when your task concerns 4. trigger a build. Begin with the documented default, test it on disposable input, and open the source section for its current options and limitations.

How to use the main workflows

Choose the smallest relevant function

Start with one capability card instead of asking the repository to do everything at once. This makes permissions, inputs, and output easier to understand.

Prepare a disposable input

Use a sample URL, copied repository, test document, or sandbox account. Keep production credentials and irreplaceable files out of the first run.

Run, observe, and stop

Watch Terminal output or the host application's activity view. If the behavior differs from the README, stop with Control + C or cancel inside the host before retrying.

Inspect the result

Check generated files, diffs, API responses, logs, or previews. Never assume “command finished” means the result is correct.

Save a repeatable recipe

Record the working command and non-secret configuration in your project's README. Store secrets in the documented environment file or password manager.

Copy-paste recipes from upstream documentation

These examples are selected from the repository's first-party documentation. Replace obvious placeholders, keep quotation marks intact, and run them only in the context indicated by the surrounding explanation.

One-Call DX

curl -X POST http://localhost:8080/api/v1/execute/async/swe-planner.build \
  -H "Content-Type: application/json" \
  -d @- <<'JSON'
{
  "input": {
    "goal": "Refactor and harden auth + billing flows",
    "repo_url": "https://github.com/user/my-project",
    "config": {
      "runtime": "claude_code",
      "models": {
        "default": "sonnet",
        "coder": "opus",
        "qa": "opus"
      },
      "enable_learning": true
    }
  }
}
JSON

Multi-Repository Mode

curl -X POST http://localhost:8080/api/v1/execute/async/swe-planner.build \
  -H "Content-Type: application/json" \
  -d '{
    "input": {
      "goal": "Add JWT auth across API and shared-lib",
      "config": {
        "repos": [
          {
            "repo_url": "https://github.com/org/main-app",
            "role": "primary"
          },
          {
            "repo_url": "https://github.com/org/shared-lib",
            "role": "dependency"
          }
        ],
        "runtime": "claude_code",
        "models": {
          "default": "sonnet"
        }
      }
    }
  }'

3. Run

af                 # starts AgentField control plane on :8080
python -m swe_af   # registers node id "swe-planner"

Configuration, accounts, and files

Configuration files

None documented

A configuration file changes behavior without changing source code. Make one change at a time and keep a backup before editing JSON, TOML, YAML, or environment files.

Important directories

None documented

Paths identify where the relevant implementation or generated files live. Paths beginning with ~ are inside your home folder.

Credentials

None documented

Prefer temporary, least-privileged credentials. Never commit .env, tokens, cookies, private keys, or session exports.

External services

model providers

Check pricing, data retention, rate limits, and account permissions before enabling optional integrations.

Safe operating habits

Use test data first and keep a current backup or Git commit.
Read commands before pasting. A README is useful evidence, not a substitute for judgment.
Review agent-generated file changes with git diff before committing.
Keep local services bound to 127.0.0.1 unless you intentionally secure and expose them.
Do not give plugins or MCP servers broader filesystem, browser, GitHub, or cloud access than their current task requires.
Confirm API costs and model names before running large batches.
Stop and investigate repeated authentication failures instead of pasting a token into multiple places.

Reference and source trail

The manual reflects first-party files checked on August 18, 2026. It explains the indexed repository rather than promising that every optional third-party integration is available or safe.