Skip to content

AgriciDaniel/agentic-habits / skills/habits

Agentic habits. Give an agent standing behavior that survives the end of a chat, in three tiers: stated rules placed where their trigger fires, gates that make the ones that must not fail deterministic, and an evidence-bound judge that scores adherence from a fresh context. Use for /habits, and whenever someone says from now on, always do this, stop doing that, remember to, make this stick, agent habits, bad habits, how do I know you actually did it, or asks why the agent keeps repeating a mistake.

assets/starter/project-habits.md

View on GitHub
<!--
MENU, NOT AN INSTALL. Four project-scope craft cards, budget is ten per project
file. Adapt every one of them to this codebase: name the real test command, the
real formatter, the real dangerous directory. A habit that names the actual tool
is followed better than a generic one. Write the result to
<repo>/.claude/rules/habits.md and replace every <added> with today's date.
-->

# Habits

Standing habits for this project. Each one is a trigger and an action. Follow
them in every session in this repository. A direct instruction from the user
outranks any habit here.

## Active

### PRJ-01 · Smallest diff
**When** I am changing code to satisfy a request.
**Do** Change only what the request needs, and propose anything else separately.
<!-- habit: id=PRJ-01 evidence=practitioner added=<added> source=starter/craft status=active check="the diff contains no unrelated renames, reformats, or refactors" -->

### PRJ-02 · Match the neighbours
**When** I write code in an existing file.
**Do** Follow that file's patterns, naming, and comment density instead of my own defaults.
<!-- habit: id=PRJ-02 evidence=practitioner added=<added> source=starter/craft status=active check="the new code is not identifiable as foreign by style alone" -->

### PRJ-03 · No green-washing
**When** a test fails and the fastest fix is to change the test.
**Instead** Fix the code, or report the failure verbatim and stop.
<!-- habit: id=PRJ-03 evidence=contested added=<added> source=starter/craft status=active check="no assertion was weakened, skipped, or deleted to produce a pass" -->

### PRJ-04 · Clean exit
**When** I am about to report a task as done.
**Do** Remove scratch files, debug output, and stray branches I created, or say what I left and why.
<!-- habit: id=PRJ-04 evidence=practitioner added=<added> source=starter/craft status=active check="nothing the agent created for its own convenience remains unmentioned" -->
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 323kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 323kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 323kUpdated Claude CodeCodexVerifiedInstall