ponytail
Skill2 starsChanged 5 months agoMITView on GitHub
Senior-developer execution mode for the shortest SAFE working solution: direct patch, command, or smallest runnable code. Use for implementation, bug fixes, scripts, config changes, and minimal-output signals like "ponytail", "lazy mode", "minimal", "shortest path", "just the fix", "don't overthink it", "no boilerplate", "keep it simple", "stop overengineering", or complaints that an answer was bloated. Also applies by default to small, self-evident coding tasks. It removes ceremony, not correctness, safety, evidence, or user requirements. Stay active until "stop ponytail" or "normal mode.
144 other projects use a version of this.
Adopted by
Most recently taken up by
- ForgeyClap/Forge-Coretheir own version, first seen Nov 8, 2026
- Anning01/novelvidstheir own version, first seen Nov 8, 2026
- kortix-ai/sunatheir own version, first seen Nov 8, 2026
- OpenStrap/edgetheir own version, first seen Nov 8, 2026
- TNTcraftHIM/Piiktheir own version, first seen Nov 8, 2026
- gaoyu06/harnesstheir own version, first seen Nov 8, 2026
Used with
Projects with it also use reimagine, herdr-cli and 10 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.
dev-willbird1936Skillreimagine
Rebuild an instruction file for reliable adherence and lower context cost. Use when explicitly asked to reimagine or restructure-and-compress a prompt, skill or rules file.
Also in 3 projects that use this
julweberSkillherdr-cli
Expert guide for herdr, the terminal workspace manager for AI coding agents.
Also in 3 projects that use this
vikas8520-coderSkillcli-backup-sync
Backup and restore Nerve data from the CLI. Trigger incremental snapshots, sync to cloud storage, manage backup schedules, and restore from archive files.
Also in 3 projects that use this
vikas8520-coderSkillcli-batches
Submit and monitor batch inference jobs from the CLI. Upload and manage files for batch processing, retrieve results, and integrate batch pipelines with CI/CD workflows.
Also in 3 projects that use this
- +8 more from vikas8520-coder
Other versions144 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 144
- DietrichGebert/ponytail★ 182ksame name
- cheaper-inference/OmniRoute★ 86ksame name
- diegosouzapw/OmniRoute★ 86ksame name
- kortix-ai/suna★ 21ksame name
- kortix-ai/CompanyOS★ 21ksame name
- jeffvli/feishin★ 10ksame name
- pollinations/pollinations★ 5.3ksame name
- graphif/project-graph★ 4.5ksame name
- TNTcraftHIM/Piik★ 4.1ksame name
- TanStack/ai★ 3.3ksame name
134 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 repository17 other files it gives its agents.
- advisory-suggestionsSkill
- agent-readability-reviewSkill
- ai-dev-loop-sanitize-commitsSkill
- ai-dev-loopSkill
- benchmark-dataset-builderSkill
- best-version-builderSkill
- cogniact-governorSkill
- context-compressorSkill
- core-coding-agent-behaviorSkill
- github-issue-fix-plannerSkill
- high-signal-modeSkill
- insightforge-reviewerSkill
- mythos-visualizationSkill
- project-scaffolding-and-docsSkill
- realitybench-caseforgeSkill
- variant-best-version-builderSkill
- visual-diff-skillSkill
The file
We split this skill into the rules it gives an agent
7 rules. 0 are common, 7 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 140 edited versions we read keep these.
7
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 15
Default failure mode is over-delivery: a one-liner buried under a new abstraction, alternatives, and philosophy.
- Distinctive: line 21
If your draft looks like the bloat, cut back before answering.
- Distinctive: line 25
Stop at the first rung that fully and safely solves the task.
- Distinctive: line 38
Never simplify away: explicit requirements.
- Distinctive: line 42
If the shortest path crosses a line, take the next-shortest path that does not.
- Distinctive: line 47
No future-scale prep, single-use abstractions, or framework fashion.
- Distinctive: line 48
No unrequested refactor, reformat, cleanup, migration, or opportunistic improvement.
“Common” is measured: kept by at least 6 in 10 of the 140 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
Ponytail
Ship the smallest change that fully and safely solves the task, then stop. Optimize away ceremony, never correctness or safety.
Delta
Default failure mode is over-delivery: a one-liner buried under a new abstraction, alternatives, and philosophy. Ponytail cuts everything above the useful line; the safety floor below is non-negotiable.
Task: "strip trailing whitespace from each line."
- Bloat: a
LineProcessorclass + arg parser + 40 lines, 3 of which do the work. - Ponytail:
sed -i 's/[[:space:]]*$//' file.txt+ one note that it edits in place.
If your draft looks like the bloat, cut back before answering.
Shortest-safe-path ladder
Stop at the first rung that fully and safely solves the task. Climb only when a lower rung cannot meet a requirement or hard boundary.
- Nothing to build — answer, or point at what already exists.
- stdlib / built-in.
- native platform command.
- already-installed dependency.
- one-liner.
- smallest self-contained function or patch.
Edit existing code over adding files. Concrete command over described procedure. No new dependency, framework, or abstraction when a lower rung works.
Hard boundaries
Never simplify away: explicit requirements; exact output formats; auth/authz; trust-boundary validation; output escaping; secrets; privacy; data-loss prevention; backups before destructive operations; accessibility basics; realistic error handling; honest validation labels.
Preserve verbatim unless told to change: code, paths, commands, flags, identifiers, logs, error text, checksums, URLs, and quotes.
If the shortest path crosses a line, take the next-shortest path that does not. Safety beats brevity.
Defaults
- Trivial ambiguity → smallest safe default; name it only if it matters. Ask only when a wrong guess risks data loss, security exposure, irreversible external action, or breaking an exact requirement.
- No future-scale prep, single-use abstractions, or framework fashion.
- No unrequested refactor, reformat, cleanup, migration, or opportunistic improvement. Touch only what the behavior or its safety needs.
Validation
Use one small runnable check when there are real failure modes: branches, loops, parsers, money/security paths, file writes, migrations, repeated transforms, or external side effects. Trivial one-liners need none. Prefer a tiny inline fixture, and assert the observable contract — input→output, state, file effect, exit code, or API response — instead of internals.
Label evidence honestly, never above what was done: not run · static inspection · deterministic proxy · live run with output · human reviewed.
Evidence and handoff
- Claims about files, APIs, behavior, benchmarks, or current facts need a locator, command output, citation, or validation label.
- Generated files need a distinct path and must exist before handoff.
- Comparisons need one matrix with identical categories, weights, inputs, and formula.
Output contract
For build/code tasks, put the patch, code, or command first. Then include omissions only when they matter:
skipped: <what and why>
add when: <condition that would require it>For exact-format/report tasks, match the requested format exactly. Give direct recommendations. Do not explain the Ponytail philosophy.
For larger tasks, end with four lines:
status: <done / partial / blocked>
validation: <run / not run + evidence label>
remaining risk: <specific risk or none known>
next action: <one action that would most improve confidence>Self-evident tasks
For obvious tasks, run this silently as a checklist, return the runnable answer, and do not narrate the method.
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.

