Coding Posture

Eleven task modes with checklists, picked before non-trivial coding work

Tested · Works

Test report

Verdict
Tested · Works
Score
9.2/10
Tested
Aug 7, 2026
Environment
Claude Code 2.x (agent harness)

Ran the same two prompts twice — a production NOT NULL migration and a 'just get the flaky test green' request — scored against a rubric written from standard DBA and CI practice, not from the skill. Without it the migration answer never mentioned a backup, a rollback path, or a staging rehearsal (2.5 of 6 rubric points); the mode checklist pulled all three in and blocked on the unanswered question of what to backfill (6 of 6). The sharper result was the flaky test: the no-skill answer volunteered pytest reruns and quarantine as 'the pragmatic move', which is precisely the fake-green the skill's always-on rules refuse. One file, no scripts, no network calls, and the copy-in install ran clean from a HOME with no .claude at all.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

  • Installs cleanly 5/5
  • Triggers reliably 5/5
  • Output vs. baseline 8/10
  • Docs & honesty 5/5

What Coding Posture does

A single SKILL.md that makes the agent name a working mode — debug, fix, review, test-first, refactor, optimize, migrate, upgrade, integrate, spike, unstuck — and follow that mode's checklist before touching code. Triggers on non-trivial coding requests such as production migrations, dependency upgrades, performance work, or a bug that has resisted several attempts. Adds always-on rules against faking green tests, reporting unrun results, and running destructive commands without scope.

How to install Coding Posture

git clone https://github.com/alexei-led/coding-posture.git
mkdir -p ~/.claude/skills
cd coding-posture && cp -r skills/coding-posture ~/.claude/skills/coding-posture

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Coding Posture

  • /coding-posture Eleven task modes with checklists, picked before non-trivial coding work

It also activates on plain-language prompts like these:

  • This flaky test keeps failing CI, just get it green so we can ship
  • Walk me through adding a NOT NULL column to the production users table
  • This bug has resisted three fixes already, help me get unstuck

Frequently asked questions

Is the Coding Posture skill free?
Yes. The skill itself is free from alexei-led/coding-posture. SkillProof publishes the install command and an independent test verdict at no cost.
Does Coding Posture work with Claude Code?
We tested it with Claude Code 2.x (agent harness) on Aug 7, 2026. Verdict: Tested · Works. Ran the same two prompts twice — a production NOT NULL migration and a 'just get the flaky test green' request — scored against a rubric written from standard DBA and CI practice, not from the skill. Without it the migration answer never mentioned a backup, a rollback path, or a staging rehearsal (2.5 of 6 rubric points); the mode checklist pulled all three in and blocked on the unanswered question of what to backfill (6 of 6). The sharper result was the flaky test: the no-skill answer volunteered pytest reruns and quarantine as 'the pragmatic move', which is precisely the fake-green the skill's always-on rules refuse. One file, no scripts, no network calls, and the copy-in install ran clean from a HOME with no .claude at all.
What is the Coding Posture SkillProof Score?
9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.
How do I install Coding Posture?
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 Coding Posture 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.