Research First

SkillDocs & knowledge

Makes your agent check current facts about libraries, APIs, and repos before planning, working like a built-in research claude skill.

Available today. Use it from your connected AI after setup.

Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.

Then ask your AI: use the Research First skill

About this skill

Codex adapter for research-before-planning work. Use when a task depends on current external facts, unfamiliar libraries, APIs, standards, security advisories, release notes, product behavior, or repo behavior that must be verified before planning or implementation.

What this skill tells your AI

The instructions your AI receives, as published by zaxbyhub/opencode-swarm in .agents/skills/research-first/SKILL.md and read by ahel’s review.

Read .claude/skills/research-first/SKILL.md for the source protocol.

Codex-specific execution notes:

  • For current or unstable external facts, browse or use primary sources and cite URLs.
  • For repo facts, prefer rg, file reads, tests, and local source over memory.
  • Separate verified facts from assumptions.
  • Do not plan implementation until the relevant unknowns are resolved or explicitly marked as risks.

When research affects OpenAI product usage, use the openai-docs skill and official sources only.

Signals

GitHub stars
478
Forks
54
Last commit
Sep 2026
Advanced
Item type
skill
Key
research-first-zaxbyhub
Source
github.com/zaxbyhub/opencode-swarm