Codex with ChatGPT Planning Loop
XiaoDuoYa/codex-with-chatgpt
Uses ChatGPT in the browser as the planning and review brain for a Codex session, with Codex keeping all execution and ChatGPT reading the workspace through a bridge.
Runs the six-agent CraftBot workflow in this specific repository, turning a design brief and references into versioned Blender Python scripts, renders, a rationale and an archived transcript.
Install Running a CraftBot Experiment
$ npx skills add lukapiskorec/craftbot --skill running-craftbot-experiment -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install lukapiskorec/craftbot running-craftbot-experiment --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/lukapiskorec/craftbot.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/running-craftbot-experiment .claude/skills/running-craftbot-experiment && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "running-craftbot-experiment" agent skill from https://github.com/lukapiskorec/craftbot/tree/main/skills/running-craftbot-experiment into .claude/skills/running-craftbot-experiment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "running-craftbot-experiment", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/lukapiskorec/craftbot/tree/main/skills/running-craftbot-experimentType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add lukapiskorec/craftbot --skill running-craftbot-experiment -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install lukapiskorec/craftbot running-craftbot-experiment --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/lukapiskorec/craftbot.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/running-craftbot-experiment .agents/skills/running-craftbot-experiment && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "running-craftbot-experiment" agent skill from https://github.com/lukapiskorec/craftbot/tree/main/skills/running-craftbot-experiment into .agents/skills/running-craftbot-experiment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "running-craftbot-experiment", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add lukapiskorec/craftbot --skill running-craftbot-experiment -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install lukapiskorec/craftbot running-craftbot-experiment --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/lukapiskorec/craftbot.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/running-craftbot-experiment .cursor/skills/running-craftbot-experiment && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "running-craftbot-experiment" agent skill from https://github.com/lukapiskorec/craftbot/tree/main/skills/running-craftbot-experiment into .cursor/skills/running-craftbot-experiment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "running-craftbot-experiment", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/lukapiskorec/craftbot.git --path skills/running-craftbot-experiment--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add lukapiskorec/craftbot --skill running-craftbot-experiment -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install lukapiskorec/craftbot running-craftbot-experiment --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/lukapiskorec/craftbot.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/running-craftbot-experiment .gemini/skills/running-craftbot-experiment && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "running-craftbot-experiment" agent skill from https://github.com/lukapiskorec/craftbot/tree/main/skills/running-craftbot-experiment into .gemini/skills/running-craftbot-experiment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "running-craftbot-experiment", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install lukapiskorec/craftbot running-craftbot-experimentInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add lukapiskorec/craftbot --skill running-craftbot-experiment -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/lukapiskorec/craftbot.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/running-craftbot-experiment .github/skills/running-craftbot-experiment && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "running-craftbot-experiment" agent skill from https://github.com/lukapiskorec/craftbot/tree/main/skills/running-craftbot-experiment into .github/skills/running-craftbot-experiment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "running-craftbot-experiment", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add lukapiskorec/craftbot --skill running-craftbot-experiment -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install lukapiskorec/craftbot running-craftbot-experiment --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/lukapiskorec/craftbot.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/running-craftbot-experiment .opencode/skills/running-craftbot-experiment && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "running-craftbot-experiment" agent skill from https://github.com/lukapiskorec/craftbot/tree/main/skills/running-craftbot-experiment into .opencode/skills/running-craftbot-experiment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "running-craftbot-experiment", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
running-craftbot-experiment- GitHub stars
- 152
- Token cost
- ~9.1k tokens
- SKILL.md length
- 4,832 words
- Files
- 1
- Skills in repo
- 13
- Repo updated
- First seen
- Licence
- Custom licence
At a glance
Runs the six-agent CraftBot workflow in this specific repository, turning a design brief and references into versioned Blender Python scripts, renders, a rationale and an archived transcript.
- Works in 6 steps: Set-up (CraftBot). Agent detected and… → Concept (Designer, Researcher).… → Build to the brief (Builders,… → …
- Starting a new CraftBot design experiment from a brief and reference materials
- SKILL.md covers Overview, Run folder and agent name, Models per tier and Roles and hand-off files, plus 13 more sections
- Calls python, gemini and claude
What it does
One run turns an experiments/NN_Name/ folder's brief and references into a building design, carried out since experiment 15 by six agents defined in .claude/agents/: CraftBot as orchestrator, plus Designer, Researcher, Builder, Inspector and Runner roles that it spawns. The user watches narration as the run proceeds and gets a full report at the end; each agent's own procedure lives in its own agent file, with modelling knowledge in separate skills and in tools/. Earlier experiments 01 to 14 were single-agent runs of the same workflow and produce the same output shape minus the hand-off files.
Every run belongs to one model and writes into a run folder named after that model, discovered by the bundled tools rather than registered anywhere else; a fixed naming table turns a reported model name like claude-opus-5-1 into a folder name like Opus 5.1 and a file slug like opus51, with Fable alone keeping its bare name for compatibility with eleven earlier experiments. The default run designs a single building; a brief that explicitly asks for several variations triggers a separate multi-variation workflow with one shared stage followed by one team per variation.
When your agent uses it
- Starting a new CraftBot design experiment from a brief and reference materials
- Resuming or redoing an existing experiment run in this repository
- Running a multi-variation CraftBot experiment that designs several versions of a building
Example prompts
- “Run the experiment in experiments/16_courtyard-house with the brief in its input folder.”
- “Resume the Opus run of experiment 12 where it left off.”
- “Run experiment 18 as a multi-variation run with three design variations.”
Requirements
- The CraftBot repository with its tools/, skills/ and .claude/agents/ setup
- Blender for the generated Python scripts and renders
Workflow steps
6 steps, taken from the first numbered list in SKILL.md.
- Set-up (CraftBot). Agent detected and /agent.md written (the folder created if missing, the agent, the mechanical model and the folder…
- Concept (Designer, Researcher). concept.md, research through one or more Researchers (each returns a verdict table; the Designer works…
- Build to the brief (Builders, Inspectors, Runner). The version loop, one fresh Builder per version: write, render, overlap and contact…
- Comparison and structural review (Designer with an Inspector). Against the reference (rationale 3b): the last model versus the photos…
- User review rounds. A user message with changes is a new phase: CraftBot appends it verbatim to the prompt file, splits it into scope (its…
- Close-out (CraftBot, Runner). Rationale compiled from the notes files, callouts written and checked, the Runner's run check, the final…
What it can do on your machine
Read from SKILL.md and the folder at commit 2c9715b. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
pythongeminiclaudecurl
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use curl, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Running a CraftBot Experiment loads about 9.1k tokens when it runs. Until then it costs about 142 tokens; SKILL.md has 4,832 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 4,832 words (~9,103 tokens).
“One run turns a brief plus reference materials into versioned Blender Python scripts, renders, a design rationale with callouts, models in the web viewer and an archived transcript. Since experiment 15 the run is done by a team of six…”
- name
- running-craftbot-experiment
Files
Just SKILL.md in skills/running-craftbot-experiment of lukapiskorec/craftbot.
Open the folder on GitHubat commit 2c9715b
Compare with similar skills
Running a CraftBot Experiment next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Running a CraftBot Experiment this skilllukapiskorec/craftbot | 152 | — | ~9.1k | Automated safety check: Pass | Custom licence | |
| Codex with ChatGPT Planning LoopXiaoDuoYa/codex-with-chatgpt | 7.1k | — | ~11k | Automated safety check: Notes | MIT | |
| CCG Tri-Model Orchestrationzereight/gitlab-mcp | 2k | 1 repos | ~657 | Automated safety check: Pass | MIT | |
| WebGPTNhahan/WebGPT | 120 | — | ~1.4k | Automated safety check: Pass | MIT | |
| Codex Workflowsscasella/claude-dynamic-workflows-codex | 325 | — | ~13k | Automated safety check: Pass | MIT | |
| Council External Codex Reviewaffaan-m/ECC | 276k | — | ~1.5k | Automated safety check: Pass | MIT |
Similar skills
CCG Tri-Model Orchestration
zereight/gitlab-mcp
Runs a task through Codex and Gemini CLIs in parallel alongside Claude, then synthesizes the three outputs into one answer with agreements and conflicts called out.
WebGPT
Nhahan/WebGPT
Hands bounded tasks from Codex to a signed-in ChatGPT web session at a chosen reasoning level, or opens a terminal chat in your project that you control.
Codex Workflows
scasella/claude-dynamic-workflows-codex
Run a dynamic-workflow script on a local Codex App Server — orchestrate many Codex / GPT agents (the agent / parallel / pipeline / phase / budget DSL) instead of Claude subagents, for codebase…
Council External Codex Review
affaan-m/ECC
Adds one optional external Codex critique that tries to break a council's decision draft, sent to OpenAI only after you consent.
Agent Squad Python Guide
2FastLabs/agent-squad
Map of the agent-squad Python framework for async multi-agent orchestration: which agent, classifier, storage and tool provider to pick, and the pitfalls to avoid.
More from lukapiskorec/craftbot
All 13 skills in this repoCraftBot Fabrication Plan Drafter
lukapiskorec/craftbot
Turns a finished CraftBot building model into cut lists, stick orders and full-scale printable sheets for building a physical scale model by hand.
Extending Previous Architectural Models
lukapiskorec/craftbot
Rules for continuing or remodeling a prior CraftBot building script: treat the inherited model as a found object, audit it against the brief, and track every change.
Modular Grids and Panelization
lukapiskorec/craftbot
Derives a building's dimensional grid from real sheet, panel, and transport constraints instead of hard-coding spacing numbers.
Non-Orthogonal Timber Geometry
lukapiskorec/craftbot
Rules for modeling sloped, twisted and warped timber structures from straight members: ruled surfaces, per-bay framing, plane-cut legs and slab-exact clipping.
Procedural Geometry for Blender Scripts
lukapiskorec/craftbot
Guides Blender Python that generates construction geometry from derived parametric expressions instead of hand-typed coordinates, so changes propagate and overlaps stay checkable.
Reading Visual References
lukapiskorec/craftbot
A skill your agent uses when the reference material for an experiment is photographs, drawings, or floor plans rather than (or alongside) a dimensioned manual, and dimensions or topology must be…
Related
Works with
Categories
Questions about Running a CraftBot Experiment
What does Running a CraftBot Experiment do?
Runs the six-agent CraftBot workflow in this specific repository, turning a design brief and references into versioned Blender Python scripts, renders, a rationale and an archived transcript. claude/agents/: CraftBot as orchestrator, plus Designer, Researcher, Builder, Inspector and Runner roles that it spawns. The user watches narration as the run proceeds and gets a full report at the end; each agent's own procedure lives in its own agent file, with modelling knowledge in separate skills and in tools/.
When should I use Running a CraftBot Experiment?
Running a CraftBot Experiment fits situations like: starting a new CraftBot design experiment from a brief and reference materials; resuming or redoing an existing experiment run in this repository; running a multi-variation CraftBot experiment that designs several versions of a building.
How do I install Running a CraftBot Experiment in Claude Code?
Run `npx skills add lukapiskorec/craftbot --skill running-craftbot-experiment -a claude-code`. Or copy the skill folder (skills/running-craftbot-experiment in lukapiskorec/craftbot) into .claude/skills/running-craftbot-experiment in your project. Claude Code loads it when a task matches its description.
How do I install Running a CraftBot Experiment in Codex?
Run `npx skills add lukapiskorec/craftbot --skill running-craftbot-experiment -a codex`. Or copy the skill folder (skills/running-craftbot-experiment in lukapiskorec/craftbot) into .agents/skills/running-craftbot-experiment in your project. Codex loads it when a task matches its description.
Can I use Running a CraftBot Experiment in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add lukapiskorec/craftbot --skill running-craftbot-experiment -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/running-craftbot-experiment, .gemini/skills/running-craftbot-experiment, .github/skills/running-craftbot-experiment and .opencode/skills/running-craftbot-experiment in your project.
What does Running a CraftBot Experiment need to run?
Going by SKILL.md and its folder, Running a CraftBot Experiment needs the command-line tools its instructions call (python, gemini, claude and curl). Our summary lists: The CraftBot repository with its tools/, skills/ and .claude/agents/ setup; Blender for the generated Python scripts and renders.
Does Running a CraftBot Experiment access the network?
SKILL.md contains no URLs. Its commands use curl, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is Running a CraftBot Experiment safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Running a CraftBot Experiment use?
Running a CraftBot Experiment has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.
How many tokens does Running a CraftBot Experiment use?
About 9.1k tokens (SKILL.md is roughly 36k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Running a CraftBot Experiment?
Skills that share tags, products or a category with Running a CraftBot Experiment: Codex with ChatGPT Planning Loop (XiaoDuoYa/codex-with-chatgpt, 7.1k stars), CCG Tri-Model Orchestration (zereight/gitlab-mcp, 2k stars), WebGPT (Nhahan/WebGPT, 120 stars) and Codex Workflows (scasella/claude-dynamic-workflows-codex, 325 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Running a CraftBot Experiment?
lukapiskorec (a GitHub user) maintains it in lukapiskorec/craftbot, which has 152 GitHub stars. The repository holds 13 skills in this directory. The repository was last updated on September 30, 2026.
Source: lukapiskorec/craftbot on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.