Ponytail lazy senior dev mode: 6 skills adapted from DietrichGebert/ponytail.
Install
# from npm (prebuilt)
dsh plugin --profile web add dsh-ponytail-skills
# from GitHub (first run asks for allowBuilds approval — follow the hint, retry)
dsh plugin --profile web add github:gongyijie85/dsh-ponytail
Any plugin you install runs third-party code with your own permissions — it can read your files, use your credentials, and reach the network, and tool approvals don’t sandbox it. GitHub-sourced plugins also run build scripts at install time — pnpm blocks those until you allow them, so an install can stop with ERR_PNPM_GIT_DEP_PREPARE_NOT_ALLOWED or ERR_PNPM_IGNORED_BUILDS; dsh prints the exact key to add under allowBuilds in your profile’s pnpm-workspace.yaml, and the install works on the next run. Allowing a build is a trust decision: only install sources you trust, and pin a commit (github:owner/repo#sha).
README
Ponytail, the lazy senior dev mode, for the DeepSeek Harness (DSH) — 6 skills adapted from DietrichGebert/ponytail (the ~76k⭐ "laziest solution that actually works" coding style).
The plugin registers a skill provider on the host layer of the ctx.skills
registry. The 6 skills ship inside the package (skills/<name>/SKILL.md); no
user configuration needed.
Unofficial port: skill content adapted from DietrichGebert/ponytail (MIT, © DietrichGebert).
Quick start
Tell the agent "use ponytail mode for this code", or let it adopt the style
on its own while coding: YAGNI → reuse existing code → standard library →
native platform features → one line → minimal code. Intensity levels:
lite / full (default) / ultra; "stop ponytail" exits the mode.
Skills
| Skill | Purpose |
|---|---|
ponytail |
Core mode: forces the laziest working solution — YAGNI → stdlib → native → one line → minimum. lite / full (default) / ultra |
ponytail-review |
Code review focused exclusively on over-engineering: one line per finding (location, what to cut, what replaces it) |
ponytail-audit |
Whole-repo over-engineering audit (repo-wide ponytail-review): ranked list, biggest cut first |
ponytail-debt |
Harvest every ponytail: comment into a debt ledger so deferrals don't rot into "later means never" |
ponytail-gain |
Show ponytail's measured impact as a compact scoreboard (benchmark medians) |
ponytail-help |
Quick-reference card for all modes, skills and commands |
Install
# npm (the plain `dsh-ponytail` name is taken by another project; this port
# publishes as dsh-ponytail-skills)
dsh plugin --profile web add dsh-ponytail-skills
# GitHub
dsh plugin --profile web add github:gongyijie85/dsh-ponytail
# Local folder (development)
dsh plugin --profile web add D:\plugins\dsh-ponytail
Restart the profile (dsh web) — the skills then load with the skill tool.
How it works
- Bundle layer —
cordis.patch.ymlinserts a plugin row over the dsh-base layer; later layers can address it by id. - Provider —
lib/index.jscallsctx.skills.registerProvider(...): scansskills/, parsesname/descriptionfrom YAML frontmatter, returns full skill definitions withresourceBasepointing at each skill directory. - Zero runtime dependencies — Node built-ins only.
Adaptation notes (vs upstream)
- Flattened descriptions: upstream's folded
description: >multi-line frontmatter was flattened to single lines — the DSH discovery parser reads scalar values only. - Skill bodies unchanged (pure prompts, no harness-specific references);
argument-hint,licenseand other metadata pass through verbatim. - Invocation: all skills are model+user invocable (upstream sets no
disable-model-invocation).
License
MIT. Skill content © DietrichGebert (ponytail); DSH port © gongyijie85. See LICENSE.
Links
More in this category
tw93/Mole#mole★ 74940
Official agent skill for the open-source Mole CLI (mo), a disk cleanup tool for macOS and Windows: read disk usage, preview and run cleanup, inspect installed apps, and read system health and cleanup history from Mole's JSON output. It previews with --dry-run and asks before removing anything, and leaves app uninstalls and privileged steps to the user's own terminal. The plugin ships only the skill; the CLI is installed separately.
zhu1090093659/dsh-web#packages/dsh-skill-explorer★ 10406
Skill center for the dsh web GUI: browse all loaded skills grouped by source, enable or disable model invocation, create new skills, and delete into a recoverable trash.
QingYunA/answer-me-with-html★ 6309
Answers a question with a one-page visual HTML explainer: the model writes a short Markdown draft and the bundled `am` CLI builds the page, with flow, sequence, tree, timeline and ER diagrams, tables and callouts.
iflytek/skillhub#dsh★ 5418
Browse and search a SkillHub registry and manage local skill installation, updates, and removal inside DeepSeek Harness. Targets DSH 0.2.0-rc.2 and 0.2.1-rc.2; not yet compatible with 0.2.1-rc.3.
hyhmrright/brooks-lint★ 1573
Provides six code quality skills for code review, architecture audits, technical debt, test quality, health dashboards, and remediation.
GanyuanRan/Aegis★ 1441
Software-engineering method pack for coding agents, with skills for baseline-first planning, systematic debugging, prompt hygiene, verification before completion, and repair/retirement tracking.
Community comments
Comments are public GitHub Discussions. Loading them connects to GitHub and Giscus; a GitHub account is required to post.