←Leaderboard/Code Quality/abide
Last commit on September 29, 2026·Created on September 18, 2026

coldteadotai/abide

“Turn repository instructions into a verifiable coding rubric.”
Combined rank
#552
across all skills
In Code Quality
#30
category rank
Stars
629
+0.8% in last 7d
Forks
48
+2.1% in last 7d
Watchers
2
0.0% in last 7d
Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
TL;DR

Checks edits against the project instructions a conventional linter cannot express. It sends each rule and diff to TypeSafe’s Jev, keeping the check independent of the agent’s conversational memory.

WHO IT'S FOR
AI agent / automation builders
enforce project rules on AI coding edits
Platform / DevEx teams
standardize AI coding conventions across repositories
Technical leads / architects
validate AI agent adherence to project guidelines
Teams sharing AI coding rules
sync project instructions with automated checkers
Repository contents

1 skill files

Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.

Use this skill

Follow the documented setup, then try a first task.

I’m using

Set up Abide for all supported agents

Installs: Machine-wide Abide hooks for every supported agent on the machine. · Claude Code, Codex, OpenCode, Pi

Before you start
  • A key accepted by Abide login.
  1. Run this complete block in a terminal. During login, choose a key type and paste your key; init then installs the agent hooks.

    npx @coldtea/abide login    # pick a key type, paste it once
    npx @coldtea/abide init     # hooks into every agent on this machine
  2. Start Claude Code, Codex, OpenCode, or Pi as usual.

README.md ↗ · Checked Nov 14, 2026

Give it something to do.

Suggested first task

Create a first inventory of project rules

Uses abide-compile

  • The repository and paths to its instruction files, such as AGENTS.md or CLAUDE.md when present.
  • Applicable scope globs for any subdirectory instruction files.
  • Optional lint configuration paths to check for overlapping enforcement.

Make sure abide-compile is installed in your agent before trying this task.

Use the abide-compile skill to read the project instruction files I identify and any pointer files they reference. Produce a first rule inventory with each rule's original wording, source path and line number, applicable directory scope, and classification as lint, model, deferred, or unenforceable. Explain deferred and unenforceable classifications, and record lint overlaps if I supply lint configuration. Use only rules from my files; stop after this inventory.
skills/abide-compile/SKILL.md ↗ · Checked Nov 14, 2026