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.
npx agentmods add skills/zhpeng24/devkit/friendly-pythonnpx skills add zhpeng24/devkit --skill friendly-pythongit clone --depth 1 https://github.com/zhpeng24/devkitWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00026 | $0.01884 |
| Opus 5 | $0.00013 | $0.00942 |
| Sonnet 5 | $0.00005 | $0.00377 |
| Haiku 4.5 | $0.00003 | $0.00188 |
Grade A, and why
friendly-python 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Friendly Python
Core Promise
Write Python that is easy to read, easy to change, easy to test, and hard to misuse.
This skill optimizes for maintainability before cleverness. Tools, types, tests, and formatting exist to protect readable code. If code is unreadable, everything else is secondary.
Treat Python as engineering code, not a demo language. Dynamic typing is not permission to pass vague shapes around. Production Python should make concepts, boundaries, and failure modes explicit.
Style north star: direct, explicit Python engineering. Prefer clear modules, intentional public surfaces, concrete names, readable imperative flow, precise errors, and minimal magic. Clarity beats ceremony.
When To Use
Use for any Python work:
- creating or changing
.py/.pyifiles - fixing bugs, type errors, lint errors, or test failures
- adding features, tests, CLI commands, packages, or project config
- reviewing Python diffs for maintainability
- deciding Python project layout, public API, or verification commands
Do not use this to impose a new toolchain on a project unless the task is explicitly project setup or cleanup.
First 60 Seconds
Before editing, build the minimum useful map:
- Identify project shape: application, service, CLI, library/SDK, script collection, notebook support, or mixed repo.
- Identify tooling:
uv.lock,poetry.lock,pdm.lock,Pipfile,tox.ini,noxfile.py,pyproject.toml,requirements*.txt. - Read local config:
pyproject.toml,pyrightconfig.json,mypy.ini,ruff.toml,pytest.ini. - Find the smallest relevant code path and tests.
- Choose the project type from
references/project-types.mdand the playbook fromreferences/task-playbooks.md.
Follow the project first. Improve quality inside the task boundary.
Task Router
| Task | Path |
|---|---|
| Bug fix | Reproduce or locate failing behavior, make the smallest readable fix, add or update behavior test, verify regression path |
| Type/lint diagnostics | Classify errors, fix source types before symptoms, avoid Any/ignore spread, run native checker |
| Feature | Define observable behavior, add minimal test or acceptance check, implement cleanly, verify public surfaces |
| Refactor/cleanup | Preserve behavior, improve names/boundaries/control flow, keep diff scoped, run existing tests |
| New Python project | Choose the smallest viable toolchain, create clear package/CLI/test structure, add lint/type/test baseline |
| Library/API change | Protect public imports, compatibility, __all__, py.typed, changelog/docs, downstream ergonomics |
| Tests | Test behavior and boundaries, not private implementation trivia; remove duplicated setup noise |
What ships with it
13 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.
- references/advanced-patterns.md 9.7 KB
- references/anti-patterns.md 3.6 KB
- references/boundary-layer-exceptions.md 6.2 KB
- references/diagnostic-workflow.md 2.3 KB
- references/examples.md 6.4 KB
- references/fix-patterns.md 8.9 KB
- references/package-layout-patterns.md 6.9 KB
- references/project-types.md 2.1 KB
- references/project-workflow.md 3.0 KB
- references/readability-contract.md 5.3 KB
- references/review-checklist.md 2.1 KB
- references/task-playbooks.md 3.3 KB
- references/tool-codes.md 9.3 KB
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.
- yesterday First seen · 185 lines · 26 tokens per session scan A 421b71a1cfe2
friendly-python is a skill published in the GitHub repository zhpeng24/devkit (2 stars, last pushed 1mo ago), licensed MIT. It adds 26 tokens to every session and 1,884 once invoked, about $0.0001 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.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…