openai-codex-agents

Build and operate OpenAI-first coding and agent workflows using Codex app/cloud, the Responses API, current GPT and Codex models, Agents SDK, hosted tools, tool search, MCP/connectors, skills, and approval-aware tool execution. Use when you need long-horizon software agents or OpenAI-native multi-agent orchestration.

Jump to install

Repository: mdbabumiamssm/LLMs-Universal-Life-Science-and-Clinical-Skills-
Last source activity: May 22, 2026 at 14:01
Detected SKILL.md language: English
Stars: 31
Forks: 8

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

A direct command skips the review prompt. Inspect the source before running it.

npx skills add https://github.com/mdbabumiamssm/LLMs-Universal-Life-Science-and-Clinical-Skills- --skill openai-codex-agents

The command stays on one line. Scroll horizontally to inspect it before copying.

Prefer a local copy? Download the files currently available to SkillsMP.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

File Explorer

3 files

SKILL.md

Source instructions · Read-only preview

Back to search Back to installation

OpenAI Codex Agents

Use this skill when the job is best served by OpenAI-native agent building rather than a generic framework wrapper.

Workflow

Start with the Responses API unless you need the Codex product surface or the Agents SDK orchestration layer.

Use Codex app, web, IDE, GitHub, or cloud tasks when the work is repository-scoped, diff-reviewable, and benefits from isolated worktrees or background execution.

Pick the model deliberately from the current model catalog; document whether the task needs a Codex-class coding model, a frontier reasoning model, a deep-research model, or a computer-use model.

Choose the tool surface explicitly: hosted web/file/code tools, hosted MCP, tool search, shell, apply_patch, computer use, or local runtime tools.

Use the Agents SDK when you need handoffs, guardrails, sessions, tracing, hosted containers, skills, shell/apply_patch control, or multi-agent composition.

Use MCP/connectors only with explicit approval policy, domain trust, OAuth/resource-discovery review, and auditability in mind.

Validate the workflow on a real repo or real task with a small eval set before wider rollout.

Guardrails

Default to approval for sensitive MCP or connector actions.

Record the exact model, reasoning setting, and tool policy used in tests.

Keep coding agents on a bounded write scope and enforce diff review before merge.

Prefer official OpenAI SDKs and docs over third-party wrappers when behavior matters.

Test for out-of-scope file edits, hidden network access, and unrelated config rewrites before granting broader autonomy.

Output Requirements

State the chosen model and why.

State whether the workflow uses Codex, Responses only, Agents SDK, or a specialized research/computer-use surface.

State the tool/approval policy and one rollback trigger.