SWE-AF: Autonomous software engineering fleet of AI agents for production-grade PRs on AgentField

Security audit verdict: SAFE · quality 66/100

None of the 42 rules matched — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against 42 published rules built on the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →

The quality score combines nine heuristic checks of the repo and its README: completeness, clarity, specificity, examples, structure, agent readiness, step-by-step content, instruction quality and security notes. It rates how well the project explains itself to an agent, not the code.

Quick Facts

About SWE-AF

SWE-AF Autonomous Engineering Team Runtime Built on AgentField Pronounced: "swee-AF" (one word) One API call → full engineering team → shipped code. Quick Start • Why SWE-AF • In Action • Factory Control • Benchmark

Compatible Skills

These tools work well together with SWE-AF for enhanced workflows:

• claude-plugins — semantic(0.34)+complementary+same_lang+similar_pop+shared_platform (57%)
• open-terminal — semantic(0.19)+complementary+same_lang+similar_pop+shared_platform (57%)
• meridian — semantic(0.17)+complementary+rare_topics+same_lang+similar_pop+shared_platform (55%)
• claude-context-local — semantic(0.17)+complementary+rare_topics+same_lang+similar_pop+shared_platform (55%)
• agent-builder — semantic(0.15)+complementary+same_lang+similar_pop+shared_platform (55%)

SWE-AF alternative? Top 6 similar tools

Looking for a SWE-AF alternative? If you're comparing SWE-AF with other codex skill tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.

• aiwg by jmagly · ⭐ 229

  Reusable project context and specialist workflows for the AI tools you already use. Agents, skills, rules and

• zeroshot by the-open-engine · ⭐ 2.1k

  Multi-agent graph orchestration engine for autonomous software development. Independent, adversarial review wi

• agent-skills by jdrhyne · ⭐ 241

  A collection of AI agent skills for Clawdbot, Claude Code, Codex

• gptme by gptme · ⭐ 4.5k

  Your agent in your terminal, equipped with local tools: writes code, uses the terminal, browses the web. Make

• pro-workflow by rohitg00 · ⭐ 2.9k

  Claude Code learns from your corrections: self-correcting memory that compounds over 50+ sessions. Context eng

• happier by happier-dev · ⭐ 2.2k

  Web, Desktop & Mobile client and orchestrator for Codex, Claude Code, OpenCode, Pi, Cursor, Grok, Antigravity,

More Codex Skill Tools

Explore other popular codex skill tools:

• humanizer ⭐ 61.8k
• refly ⭐ 7.5k
• nano-banana-pro-prompts-recommend-skill ⭐ 1.9k
• pi-skills ⭐ 2.5k
• skillkit ⭐ 1.5k
• upskill ⭐ 762
• Claude-to-IM-skill ⭐ 2.9k
• learning-opportunities ⭐ 2.4k
• solana-dev-skill ⭐ 584
• Skills-Manager ⭐ 991

Popular Go Agent Tools

• github-mcp-server ⭐ 34.3k · MCP Server
• nginx-ui ⭐ 11.7k · MCP Server
• xiaohongshu-mcp ⭐ 17.0k · MCP Server
• httprunner ⭐ 4.3k · MCP Server
• ENScan_GO ⭐ 4.6k · MCP Server

How this security grade is produced

Grades come from a rule-based scan built on the SlowMist agent-security taxonomy: 42 published rules, covering credential harvesting, data exfiltration, curl | sh installers and more. It is a first-layer scan, not a manual audit — we say so rather than overstate it.

The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.

Sources & who's responsible:

• SlowMist agent-security taxonomy
• Liu et al., An Empirical Study of Agent Skill Security (2026)
• Our census dataset (DOI 10.5281/zenodo.21292799) · Hugging Face mirror
• How we grade: all 42 rules, the source tiers and known limits
• Full methodology & who runs this