briefbound-ai-research-loop

briefbound-ai-research-loop is a skill for Claude Code, Codex from CCDawn/codex-skills. It costs 56 tokens per session (1,932 once invoked), scanned A, original, MIT.

A repeatable workflow for AI and machine-learning research, from reproducing a baseline and testing a specific idea to evaluating results and choosing the next direction.

In plain words
What is it for?
Use it to reproduce papers or repositories, run controlled experiments and ablations, combine findings across trials, recover from stalled progress, and hand off reproducible results.
Why use it?
It replaces unstructured trial-and-error with recorded experiments, clear comparisons, and evidence-based decisions while guarding against changing the data split or evaluation rules.

Skill for Claude CodeCodex

Written for Claude Code and Codex: shipped in a Claude Code plugin, but also agents/openai.yaml present.

Part of the briefbound-skills plugin — 30 skills shipped together

Good fit Use it to reproduce papers or repositories, run controlled experiments and ablations, combine findings across trials, recover from stalled progress, and hand off reproducible results.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/ccdawn/codex-skills/briefbound-ai-research-loop
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add CCDawn/codex-skills --skill briefbound-ai-research-loop
Clone the repo
git clone --depth 1 https://github.com/CCDawn/codex-skills

Made for: Claude Code, Codex.

Or install briefbound-skills, the plugin that ships this one along with the rest of its 30 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for briefbound-ai-research-loop

README.md
[![agentmods](https://agentmods.dev/badge/skills/ccdawn/codex-skills/briefbound-ai-research-loop/github.svg)](https://agentmods.dev/skills/ccdawn/codex-skills/briefbound-ai-research-loop)
Your own site
<a href="https://agentmods.dev/skills/ccdawn/codex-skills/briefbound-ai-research-loop"><img src="https://agentmods.dev/badge/skills/ccdawn/codex-skills/briefbound-ai-research-loop/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for briefbound-ai-research-loop

Your own site · 80×15
<a href="https://agentmods.dev/skills/ccdawn/codex-skills/briefbound-ai-research-loop"><img src="https://agentmods.dev/badge/skills/ccdawn/codex-skills/briefbound-ai-research-loop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 56 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,932 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00056 $0.01932
Opus 5 $0.00028 $0.00966
Sonnet 5 $0.00011 $0.00386
Haiku 4.5 $0.00006 $0.00193

Measured 9d ago against content hash 715701a4e762, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

briefbound-ai-research-loop scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/research/briefbound-ai-research-loop/SKILL.md · 124 lines

How it starts

The opening of the file, as written. The whole thing — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AI 研究工程循环

目标

把 AI 研究从“不断尝试代码”变成两个相互连接的轻量循环:

内层:可信 baseline -> 可证伪假设 -> 最小实验 -> 评估 -> 接受/拒绝/保留
外层:汇总多轮证据 -> 提炼规律 -> 更新假设组合 -> 继续/分支/转向/停止

本 skill 是 AI 研究工程的主 owner。briefbound-score-loop 只承接一条可量化实验 lane;竞赛规则、提交和 leaderboard 全生命周期仍由 briefbound-competition-research-lifecycle 适配。

Briefbound task contract

  • Context Boundary: 研究问题、代码和数据来源、active baseline、评价协议、允许修改面、计算预算、实验记录和当前证据。
  • Output Contract: baseline 复现结论、假设组合、实验 lane、证据综合、研究方向决策或可复现交接。
  • Allowed Action: 在已锁定的可编辑面和预算内复现、修改、运行、评估和记录;不静默改变数据划分、metric、baseline 或研究目标。
  • Success Evidence: 可复现命令、baseline 指纹、metric 与方差、diff/config、实验 artifact、对照/消融结果以及有来源的研究结论。
  • Stop Condition: baseline 不可信、评价协议漂移、数据泄漏、预算或权限不足、结果不可复现、关键假设无法区分,或继续实验已无新的信息价值。
  • Route Out: briefbound-score-loopbriefbound-feature-reuse-researchbriefbound-bug-reviewbriefbound-research-rigor-reviewbriefbound-competition-research-lifecycle、完成交接或 BLOCKED。

统一调用契约

  • 只处理 Briefbound task contract 范围;不匹配时回 briefbound-router 或更具体 owner,复合任务不吞其他 owner。
  • 用户可见内容默认中文,完成只报状态、产出、证据和剩余风险;代码、命令、路径、错误原文、API/协议、skill 名和枚举保留原样;Route Out 仅以 Briefbound task contract 为准,末行写 下一步建议: <一个具体动作>

所有权判断

  • 用户要推进一个 AI/ML 研究问题、复现论文、做消融或从多轮实验中决定方向:本 skill 主责。
  • 用户已经给出明确 baseline、metric 和单个低成本候选:本 owner 可直接比较;只有反复晋升、榜单反馈或持久 score lane 才路由 briefbound-score-loop
  • 主要问题是训练脚本、metric、数据 schema、seed、shape、NaN 或环境的确定性故障:临时路由 briefbound-bug-review,修复后返回研究循环。
  • 主要问题是 Kaggle、竞赛规则、提交包或 public leaderboard:由竞赛生命周期主责,本 skill 只承接其研究阶段。
  • 需要搜索论文、仓库、模型或可复用实现且结果会改变方案:使用 briefbound-feature-reuse-research;研究 owner 保留方向决策权。

启动快照

先从仓库、论文、日志和配置读取已有事实,只补会改变研究决策的缺口:

  • 研究问题与可观察成功标准;
  • baseline 来源、版本、命令和已知结果;
  • 数据版本、split、metric、seed 与评估预算;
  • 可编辑面、保护面和算力/时间限制;
  • 已尝试方向、失败证据和当前最可信结论。

缺少正式工件时可先运行可逆 probe,不因模板不全阻塞探索。只有 baseline、metric 或数据边界不清会让实验失去解释性时才暂停询问。

自适应流程

按当前研究不确定性选择最低充分重量:

  • QUICK:可信 baseline 上的 1-2 个低成本假设;内部维护短记录,直接实验和汇报。
  • STANDARD:需要多轮消融、多个候选或跨会话延续;维护紧凑研究契约和 append-only 实验记录。
  • DEEP:高成本训练、结论将用于论文/发布、数据或评价风险高;增加协议冻结、复现检查、严谨性审查和明确停止预算。

Read the full file on GitHub · 124 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 124 lines · 56 tokens per session scan A 715701a4e762

Subscribe to this mod's changes

briefbound-ai-research-loop is a skill published in the GitHub repository CCDawn/codex-skills (4 stars, last pushed 26d ago), licensed MIT. It adds 56 tokens to every session and 1,932 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

chinese-git-workflow

A reference for configuring Git with Chinese code-hosting services such as Gitee, Coding.net, GitLab China, and CNB, including SSH, HTTPS, credentials, CI, and repository mirroring.

jnMetaCode/superpowers-zh · 69 tokens

chinese-documentation

A Chinese technical-documentation style guide covering spacing, punctuation, numbers, terminology, and links when Chinese and English appear together.

jnMetaCode/superpowers-zh · 62 tokens

chinese-commit-conventions

A Chinese-language guide to Conventional Commits, a format for writing consistent Git commit messages, plus related changelog, commit-checking, and commit-helper configuration.

jnMetaCode/superpowers-zh · 65 tokens

systematic-debugging

A step-by-step method for finding the underlying cause of technical problems before changing code. It covers reading errors, reproducing failures, checking recent changes, and tracing data across system components.

jnMetaCode/superpowers-zh · 24 tokens

workflow-runner

A procedure for running agency-orchestrator YAML workflows directly inside Claude Code, OpenClaw, or Cursor. The current AI session performs each named role, so no separate API key is needed.

jnMetaCode/superpowers-zh · 61 tokens

seedance-vocab-zh

This skill should be used when the user asks for Chinese Seedance 2.0 prompt wording, Mandarin cinematic vocabulary, Chinese prompt compression, or translation of camera, lighting, action, VFX, audio, and production terms into Chinese.

Emily2040/seedance-2.0 · 55 tokens