claude

habits

Standing instructions for an AI agent as one markdown file each: habits (rules, always on or triggered) and SOPs (procedures). Installs one small generated block in CLAUDE.md/AGENTS.md pointing at a private ~/.claude-user-data folder. Ships a library of example habits to adopt, and skills to add, adopt, update, remove, list, check and share habits, write and run SOPs, sync, and verify the prompt wiring.

2 on GitHubby Daniel Rosehillv1.1.0MITupdated 10 days ago

Not in a marketplace yet, so it loads for one session with --plugin-dir.

Plugin habits in danielrosehill/Agent-Habits. Commands read from the repo's manifest — 9 days ago.

What it installs

Skills (17)

  • add-habit

    Add a new habit to your own habits — a standing rule for how the agent should behave across all work, or whenever a stated trigger applies — with a one-line summary that sits in every session. Writes ~/.claude-user-data/habits/<id>.md, reinstalls the CLAUDE.md block and syncs. Use when the user stat

  • adopt-habit

    Adopt a habit from the example library into your own habits — copies it to ~/.claude-user-data/habits/, fills in tunable values such as a spend limit, records which library version it came from, then reinstalls the CLAUDE.md block. Use when the user wants one of the example habits, asks "what habits

  • check-habits

    Validate your habits and SOPs — they parse, summaries are present and faithful to their full text, no two habits overlap or contradict, nothing is filed as the wrong kind, the user-data repo is committed, and adopted habits are up to date with the library. Use after editing habits or SOPs, before tr

  • context-check

    Verify the user-level CLAUDE.md is wired correctly to your habits and SOPs — exactly one generated agent-habits block, in index mode, under its size budget, matching the current habit and SOP files, with no legacy or hand-written duplicates — and repair drift by reinstalling. Use when asked to "chec

  • install-habits

    Install or update the generated agent-habits block — your habits' summary lines and SOP trigger lines — in ~/CLAUDE.md, or put full habit texts (a subset, inline) into a project CLAUDE.md or AGENTS.md that others read. Marker-delimited, idempotent, backs the file up, and leaves hand-written prose al

  • list-habits

    List your habits — ids and one-line rules, optionally retired ones — or the example library you can adopt from, or find which habit covers something. Use when asked "what are my habits", "list habits", "what habits are available", "which habit covers X", or before deciding whether something is alrea

  • reconcile-habits

    Compare a system prompt file against your habits and SOPs and resolve the drift in both directions — prose in ~/CLAUDE.md that should become a habit file, habit files whose installed copy is stale, and hand-written sections now duplicated by the managed block. Produces a reconciliation report and ap

  • remove-habit

    Retire one of your habits so it stops being installed, or delete it outright when explicitly asked — then reinstall the CLAUDE.md block and sync. Use when the user says to drop, remove, delete, retire or stop applying a habit.

  • setup

    First-time setup of Agent-Habits — create the private user-data folder for your own habits and SOPs, optionally adopt habits from the library, put it under git with a private remote, and add the one generated block to the user-level CLAUDE.md. Also covers a second machine (clone the existing user-da

  • share-habit

    Contribute one of your own habits to the public example library in the Agent-Habits plugin repo — depersonalised, with tunable values turned into variables — so others can adopt it. Use when the user says a habit might be useful to others, asks to publish or share a habit, or after add-habit produce

  • sop-edit

    Amend an existing SOP — fix a step that has drifted, add a failure mode that was just hit, adjust its trigger or autonomy, or re-date last_verified after a successful run. Use when the user says an SOP is out of date or wrong, when following one produced a surprise, and routinely at the end of any S

  • sop-list

    Show what is in the SOP library — everything, or filtered by tag, or only the procedures that have gone stale. Use when the user asks what SOPs they have, whether one exists for something, which ones need reviewing, or wants an overview of their procedures before deciding what to write next.

  • sop-remove

    Retire an SOP that no longer applies — moving it to the library's archive, or deleting it outright when asked. Use when the user says a procedure is obsolete, that they do not do it that way any more, or that a system it depends on is gone; and when a replacement SOP has been written for the same jo

  • sop-run

    Find the user's standard operating procedure for the task at hand and follow it. Reads the SOP index, matches the situation, opens exactly one procedure and executes it at the autonomy level that SOP declares. Use whenever a multi-step recurring task comes up — deploying, publishing, filing, orderin

  • sop-write

    Capture a recurring procedure as a new SOP in the user's library — usually written from what just happened in this session, while the details are still known. Use when the user says "write this up as an SOP", "remember how we did this", "add this to my procedures", or when the same multi-step sequen

  • sync

    Commit and push your habits and SOPs (the ~/.claude-user-data repo), or pull them onto this machine, then reinstall the CLAUDE.md block so it matches. Use when the user asks to sync, save, push or back up habits or SOPs, when setting up a second machine, and at the end of any session that added or c

  • update-habit

    Revise one of your habits — its rule, summary, trigger, exceptions or values — then reinstall the CLAUDE.md block and sync. Use when the user corrects how a standing habit should work, sharpens or loosens one, or says a habit's one-line summary misleads. Also for pulling in a newer library version o

Slash commands (1)

  • /habits:sop

    Run one of your standard operating procedures — or list, write, edit or retire them

View source on GitHub

Similar plugins

data-agent-kit

claudecodex

This plugin provides a specialized suite of skills for data engineers and database practitioners working on Google Cloud. It acts as an expert assistant, allowing you to use natural language prompts in your preferred coding agent to architect complex data pipelines, transform data with dbt, write Spark and BigQuery SQL notebooks, and orchestrate end-to-end workflows across GCP's data ecosystem.

37 skills

anthropics/claude-plugins-official 47,076

Turn any learning intent into a personalized Coursera experience. Asks three quick questions (topic, familiarity, preferred format), searches Coursera's catalog, and delivers the right next step — a course, hands-on project, short video, or live roleplay — then maps a path forward. Requires the Coursera connector for catalog tools.

1 skill

anthropics/knowledge-work-plugins 35,601

Internal BizOps domain. v2.8.0 ships 7 skills: orchestrator + process-mapper (BPMN/bottleneck/cycle-time, Lean+TOC) + vendor-management (scorecard+SLA+3rd-party risk, NIST SP 800-161/ISO 27036) + capacity-planner (Erlang-C queueing math for ops teams, NOT engineering) + internal-comms (ADKAR+Kotter 8-step, NOT marketing) + knowledge-ops (SOP+runbook+KB hygiene with 5W2H, context: fork) + procurement-optimizer (UNSPSC spend categorization + supplier consolidation). Orchestrator uses context: fork to route inquiries via Matt Pocock grill discipline (one question per turn, recommended answer, canon-cited challenge). Every SKILL.md ships a Forcing-question library section. 18 stdlib Python tools, 24+ reference docs. Distinct from business-growth (external sales) and c-level-advisor (strategic).

7 skills · 8 commands · 1 agent

alirezarezvani/claude-skills 29,725

data-agent-kit

claudecodex

This plugin provides a specialized suite of skills for data engineers and database practitioners working on Google Cloud. It acts as an expert assistant, allowing you to use natural language prompts in your preferred coding agent to architect complex data pipelines, transform data with dbt, write Spark and BigQuery SQL notebooks, and orchestrate end-to-end workflows across GCP's data ecosystem.

37 skills

google/skills 24,091

RLHF and preference alignment including TRL, GRPO, OpenRLHF, SimPO, verl, slime, miles, and torchforge. Use when aligning models with human preferences, training reward models, or large-scale RL training.

98 skills

Orchestra-Research/AI-Research-SKILLs 14,502

Use when the user wants Claude Code to remember their preferences across sessions, asks for a memory mod or memory harness for Claude Code, mentions the Jev memory classifier or a "classifier on every turn" mod, or wants to build a Claude Code mod on treg and Jev. Installs a ready-made Claude Code mod that has Jev judge each prompt after the turn and saves lasting preferences to .claude/jev-memory.md, or builds the same mod from one prompt.

1 skill

superdesigndev/treg 7,890

Maintain habits?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex pluginslisted