Night Market Build And Env
Bootstraps a claude-night-market dev environment: uv-first setup, Python tiers, known traps
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 31, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · f6cb297
Cloned the repo to /tmp and ran the skill's own verification sequence: `make validate-all` printed 22 blocks, 22 "OK Plugin validation passed", 0 "(validation failed)" — output shape exactly as documented (skill says 26 directories; actual is 22), and `make -C plugins/leyline test` ended with "719 passed, 6 warnings in 8.83s" plus a coverage table (skill's snapshot said 710 passed / 9.96s). For the output test I wrote a BASELINE ONBOARDING.md from only README.md "Requirements" + `make help` (what a no-skill agent sees) and a SKILL ONBOARDING.md following SKILL.md, then executed the two places they diverge: baseline's `pytest` collected 1502 tests with ZERO from plugins/ (leyline's 719 absent — a false green the skill's `norecursedirs` trap warns about, confirmed at pyproject.toml:52), and baseline's `python3 -m venv && pip install -e .` produced a venv containing pytest but no ruff/mypy/pre-commit/bandit (they sit in the [dev] extra), while README never mentions uv even though the Makefile calls `uv run` 18 times. Against that: the skill is a v1.9.15 snapshot and HEAD is 1.9.17, so its per-plugin Python-floor list is now wrong (all 23 plugins pin ">=3.9"), root ruff target-version is "py39" not "py312", setup-uv is pinned v9.0.0 not v8.2.0, uv.lock has ruff 0.15.22 not 0.14.13, and the root package.json it cites for the @google/gemini-cli pin returns 404 — hence install 4 and docs 3. Spot-checked scripts/check_pinned_versions.py, scripts/check_ruff_version.py, .github/workflows/python39-compat.yml: all HTTP 200, and the workflow really does contain both described gates (UP007 ruff check + `uv venv --python 3.9` with `--override-ini="addopts="`). No curl|sh, base64, secrets exfil, or injection text; install verified in a throwaway HOME.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 4/5
- Triggers reliably 5/5
- Output vs. baseline 7/10
- Docs & honesty 3/5
What Night Market Build And Env does
A repo-specific onboarding runbook for the claude-night-market monorepo: an ordered toolchain table (uv, Python 3.12, make, ruff/mypy/bandit/pre-commit via uv.lock), a four-step first-hour verification sequence with the exact output shape each step should print, the three Python version tiers (root 3.12 tooling vs 3.9-constrained hook import chains), and a table of traps that have broken the environment before. Triggers when someone is setting up the repo from a bare machine, when the toolchain breaks (wrong ruff on PATH, pre-commit bandit env failing), or when asking why hook code is pinned to Python 3.9 while root tooling needs 3.12. Not for daily test/lint/release commands — it redirects those to night-market-operations.
How to install Night Market Build And Env
git clone --depth 1 https://github.com/athola/claude-night-market.git /tmp/night-market-build-and-env-src
mkdir -p ~/.claude/skills
cp -R /tmp/night-market-build-and-env-src/.claude/skills/night-market-build-and-env ~/.claude/skills/night-market-build-and-env
# Single self-contained SKILL.md, no bundled scripts or assets.
# No API keys or external deps needed to READ the skill.
# To actually FOLLOW it you need: uv (astral.sh/uv), Python 3.12, make + bash,
# and a clone of athola/claude-night-market. Rust/node/mdbook/gh are optional.
# Scope note: this skill only describes the claude-night-market repo itself.
# It is not a generic uv/Python setup guide for other projects.
# Marketplace alternative: /plugin marketplace add athola/claude-night-market
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Night Market Build And Env
-
/night-market-build-and-envBootstraps a claude-night-market dev environment: uv-first setup, Python tiers, known traps
It also activates on plain-language prompts like these:
-
Rebuild my dev environment from scratch -
Fix a broken uv toolchain setup -
Onboard me onto this Python project's environment
Frequently asked questions
- Is the Night Market Build And Env skill free?
- Yes. The skill itself is free from athola/claude-night-market. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Night Market Build And Env work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 31, 2026. Verdict: Tested · Works. Cloned the repo to /tmp and ran the skill's own verification sequence: `make validate-all` printed 22 blocks, 22 "OK Plugin validation passed", 0 "(validation failed)" — output shape exactly as documented (skill says 26 directories; actual is 22), and `make -C plugins/leyline test` ended with "719 passed, 6 warnings in 8.83s" plus a coverage table (skill's snapshot said 710 passed / 9.96s). For the output test I wrote a BASELINE ONBOARDING.md from only README.md "Requirements" + `make help` (what a no-skill agent sees) and a SKILL ONBOARDING.md following SKILL.md, then executed the two places they diverge: baseline's `pytest` collected 1502 tests with ZERO from plugins/ (leyline's 719 absent — a false green the skill's `norecursedirs` trap warns about, confirmed at pyproject.toml:52), and baseline's `python3 -m venv && pip install -e .` produced a venv containing pytest but no ruff/mypy/pre-commit/bandit (they sit in the [dev] extra), while README never mentions uv even though the Makefile calls `uv run` 18 times. Against that: the skill is a v1.9.15 snapshot and HEAD is 1.9.17, so its per-plugin Python-floor list is now wrong (all 23 plugins pin ">=3.9"), root ruff target-version is "py39" not "py312", setup-uv is pinned v9.0.0 not v8.2.0, uv.lock has ruff 0.15.22 not 0.14.13, and the root package.json it cites for the @google/gemini-cli pin returns 404 — hence install 4 and docs 3. Spot-checked scripts/check_pinned_versions.py, scripts/check_ruff_version.py, .github/workflows/python39-compat.yml: all HTTP 200, and the workflow really does contain both described gates (UP007 ruff check + `uv venv --python 3.9` with `--override-ini="addopts="`). No curl|sh, base64, secrets exfil, or injection text; install verified in a throwaway HOME.
- What is the Night Market Build And Env SkillProof Score?
- 7.6/10 — installs cleanly 4/5, triggers reliably 5/5, output vs. baseline 7/10, docs & honesty 3/5.
- How do I install Night Market Build And Env?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Night Market Build And Env with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.