rubber-duck
Skill4 starsChanged 57 days agoApache-2.0View on GitHub
Think through a design, feature, or problem in conversation before any code or spec exists, and record the result as an idea file. Use for /r-d, /rubber-duck, ideation, design discussion, or working out an approach.
24 other projects use a version of this.
Adopted by
Most recently taken up by
- jingyibuilds/rubber-ducktheir own version, first seen Nov 15, 2026
- draekien/skillstheir own version, first seen Nov 15, 2026
- Osipchuk/agent-skillstheir own version, first seen Nov 8, 2026
- jsleemaster/gooblintheir own version, first seen Nov 1, 2026
- vidluther/skillstheir own version, first seen Oct 31, 2026
- damacus/agent-skillstheir own version, first seen Oct 29, 2026
Used with
Projects with it also use caveman, grill-me and 4 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
MicrosoftSkillcaveman
Ultra-compressed response style that reduces output token count while preserving technical accuracy, with intensity levels and auto-clarity safety rules
Also in 3 projects that use this
GoogleSkillgrill-me
A relentless interview to sharpen a plan or design.
Also in 5 projects that use this
MicrosoftSkillhandoff
Compact the current conversation into a handoff document for another agent to pick up.
Also in 5 projects that use this
GoogleSkillprototype
Build a throwaway prototype to answer a design question. Use when the user wants to sanity-check whether a state model or logic feels right, or explore what a UI should look like.
Also in 3 projects that use this
- +2 more from google
Other versions24 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 24
- kirodotdev/KiroCrew★ 4.8ksame name
- thoughtbot/rails-consultant★ 29same name
- jim60105/copilot-prompt★ 22same name
- stevepolitodesign/skills★ 10same name
- The-Artificer-of-Ciphers-LLC/skills-from-the-artificer★ 6same name
- jmcelreavey/devhub★ 5same name
- oneafrikan/the-grid★ 5same name
- Osipchuk/agent-skills★ 5same name
- dfrysinger/skills★ 3same name
- jsleemaster/gooblin★ 3same name
14 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository14 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- commits-and-prsSkill
- decomposeSkill
- github-boardSkill
- implementSkill
- releaseSkill
- reviewSkill
- reviseSkill
- specifySkill
- thunderstormSkill
- triageSkill
- worktreeSkill
- writing-docsSkill
The file
We split this skill into the rules it gives an agent
5 rules. 0 are common, 4 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 24 edited versions we read keep these.
4
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 16
Check claims against the repository before repeating them.
- Distinctive: line 19
When a decision rests on something unknown, find out rather than assume.
- Distinctive: line 22
Do not restate what was already settled.
- Very specific: line 66
Use the writing-docs skill for the prose.
- Distinctive: line 77
Create branches, worktrees, or pull requests.
“Common” is measured: kept by at least 6 in 10 of the 24 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Rubber duck
Design discussion. No implementation, no branch, no pull request.
How to talk
Argue with the idea. Agreement that arrives without resistance is not useful here.
- Say what you would do and why, then say what would change your mind.
- Check claims against the repository before repeating them. A wrong fact stated confidently costs more than a slow answer.
- Name the tradeoff instead of listing options. Recommend one.
- When a decision rests on something unknown, find out rather than assume. Read the code, run the command, check the branch.
- Say when a premise is wrong, including a premise the user supplied.
- Do not restate what was already settled. Move to what is not.
Research claims that carry weight. Cite the source and its confidence. A practitioner blog post and a documented API limit are not the same kind of evidence, and the difference belongs in the answer.
What to write down
The conversation is not the artifact. When the discussion settles, or when the
user asks, write an entry to internal/ideas/.
python3 .claude/scripts/ulid.py---
name: <short-kebab-case-name>
last_updated: <YYYY-MM-DD>
id: <ULID>
---
# <Title>
## Problem
What is wrong now, with evidence.
## Decisions
What was settled, and why. One line each.
## Open
What is not settled, and what would settle it.
## Sources
Links, with what each supports.Record decisions, not the discussion that produced them. Keep the reasoning that would be expensive to reconstruct and drop the rest.
Use the writing-docs skill for the prose.
Moving on
An idea becomes work when it has acceptance criteria someone could verify. At that point file it as an issue, or as an issue with sub-issues, and link the idea file. Until then it stays an idea.
Do not
- Write implementation code.
- Create branches, worktrees, or pull requests.
- File issues without being asked.
- Expand the idea file into a specification. A spec is a separate document with a separate review.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

