toppingout
llms.txt14 starsChanged 20 days agoMITView on GitHub
A spec-driven software factory for Claude Code that works through your tickets overnight, with a planner/executor split: a planner writes each ticket's spec in a fresh session, a cheaper model does the coding, another model family reviews it, and your Claude Code limit lasts far longer. Each job ends with evidence you can check: what was asked, what was built, what was tested, and what reviewers found. Topping Out is open source (MIT): https://github.com/kaithoughtarchitect/toppingout. Every skill and agent page has…
In this repository32 other files it gives its agents.
- AGENTS.mdAGENTS.md
- docs/llms-full.txtllms.txt
- adjudicateSkill
- agent-councilSkill
- converseSkill
- f-archive-searchSkill
- f-archiveSkill
- f-blueprintSkill
- f-boardSkill
- f-cardSkill
- f-catchupSkill
- f-complySkill
- f-contextSkill
- f-distillSkill
- f-dreamSkill
- f-greploopSkill
- f-inspectSkill
- f-night-shiftSkill
- f-plancheckSkill
- f-scheduleSkill
- f-searchSkill
- f-setupSkill
- f-shiftSkill
- f-signoffSkill
- f-sketchSkill
- f-statusSkill
- f-surveySkill
- model-councilSkill
- orca-pi-connectSkill
- pack-reviewSkill
- speakSkill
- xreviewSkill
The file
Files
# Topping Out > A spec-driven software factory for Claude Code that works through your tickets overnight, with a planner/executor split: a planner writes each ticket's spec in a fresh session, a cheaper model does the coding, another model family reviews it, and your Claude Code limit lasts far longer. Each job ends with evidence you can check: what was asked, what was built, what was tested, and what reviewers found. Topping Out is open source (MIT): https://github.com/kaithoughtarchitect/toppingout. Every skill and agent page has a Markdown twin, linked below. ## Pages - [Topping Out](https://toppingout.dev/): Open-source software factory for Claude Code: a planner specs each ticket, a cheaper model builds it, another model family reviews it. Runs overnight. - [Playbooks](https://toppingout.dev/playbooks.html): Who does what: the foreman's, the builder's and your job, and the playbook each role follows. - [Order of works](https://toppingout.dev/route.html): Every step from an idea to a merged job: the 11 steps, each opening into exactly what happens. - [Your repo](https://toppingout.dev/repo.html): A site plan of every folder Topping Out adds to your repo, and where one job's files appear. - [Logbook](https://toppingout.dev/logbook.html): The Logbook: the cards Topping Out keeps, how a lesson becomes a card and then a rule, its skills and its files. - [Skills and agents](https://toppingout.dev/skills.html): Every Topping Out skill and agent: 30 skills and 14 agents. - [Skill Atlas](https://toppingout.dev/atlas.html): Every Topping Out skill and agent as a map, with the links between them. - [Example board](https://toppingout.dev/board.html): What a Topping Out project board looks like, and how to read it: a made-up project with synthetic data. - [Requirements](https://toppingout.dev/requirements.html): The setup Topping Out is built and run on, as a site induction: tick off the kit you have, and the pass hands you the install lines. - [Cloud setup](https://toppingout.dev/cloud.html): Set up your own cloud software factory: Topping Out on a small Linux server, driven through Orca, while your computer sleeps. - [About](https://toppingout.dev/about.html): Kai Thought Architect, who drew Topping Out, and the systems it grew from. ## Skills and agents (Markdown) - [All skills and agents](https://toppingout.dev/skills.md): the full list, one line each - [code-explorer](https://toppingout.dev/agents/code-explorer.md): Trace how an existing feature works through the code. - [handoff-builder](https://toppingout.dev/agents/handoff-builder.md): Write the handover notes so the next session can pick up the work. - [lens-code](https://toppingout.dev/agents/lens-code.md): Look for bugs and broken project rules in the changed code. - [lens-comments](https://toppingout.dev/agents/lens-comments.md): Check that comments and docs still match the code. - [lens-errors](https://toppingout.dev/agents/lens-errors.md): Find errors that get silently ignored, and risky fallbacks. - [lens-instructions](https://toppingout.dev/agents/lens-instructions.md): Check prompts, skills and templates for contradictions and broken references. - [lens-loose-ends](https://toppingout.dev/agents/lens-loose-ends.md): Find what a change should also have touched: an untouched copy, a sibling line, a companion piece, an existing function. - [lens-migrations](https://toppingout.dev/agents/lens-migrations.md): Review database changes: migrations, queries and access rules. - [lens-simplify](https://toppingout.dev/agents/lens-simplify.md): Spot code that could be simpler, without changing what it does. - [lens-tests](https://toppingout.dev/agents/lens-tests.md): Find changes that aren't properly tested. - [lens-types](https://toppingout.dev/agents/lens-types.md): Check that the data types can't hold invalid values. - [plan-validator](https://toppingout.dev/agents/plan-validator.md): Check a plan against the real code before it's built (used inside f-plancheck). - [simplicity-critic](https://toppingout.dev/agents/simplicity-critic.md): Find the overbuilt parts of a plan and suggest the simplest version. - [spec-validator-fixer](https://toppingout.dev/agents/spec-validator-fixer.md): Correct statements in a plan that are provably wrong. - [adjudicate](https://toppingout.dev/skills/adjudicate.md): Go through a review point by point: accept, reject, partly accept, or can't verify. - [agent-council](https://toppingout.dev/skills/agent-council.md): Ask AI agents that can read the code to help settle a design choice. - [converse](https://toppingout.dev/skills/converse.md): Read every reply out loud until you turn it off. - [f-archive-search](https://toppingout.dev/skills/f-archive-search.md): Search the archived Logbook cards. - [f-archive](https://toppingout.dev/skills/f-archive.md): Move old Logbook cards to the archive when there are too many, keeping a summary set. - [f-blueprint](https://toppingout.dev/skills/f-blueprint.md): Write the full plan once every decision is made. - [f-board](https://toppingout.dev/skills/f-board.md): Update the project board when a job starts, merges, gets stuck or is added. - [f-card](https://toppingout.dev/skills/f-card.md): Save one lesson worth keeping as a Logbook card. - [f-catchup](https://toppingout.dev/skills/f-catchup.md): Write Logbook cards for past jobs that never got any. - [f-comply](https://toppingout.dev/skills/f-comply.md): Check whether the agent really follows a rule when it's tempted to break it. - [f-context](https://toppingout.dev/skills/f-context.md): Get a quick, complete picture of a repo before working in it. - [f-distill](https://toppingout.dev/skills/f-distill.md): Turn the most important lessons into firm rules, with your approval. - [f-dream](https://toppingout.dev/skills/f-dream.md): Look back over past jobs: merge repeated lessons and spot mistakes that keep happening. - [f-greploop](https://toppingout.dev/skills/f-greploop.md): Run an open pull request through Greptile's review, up to three rounds. - [f-inspect](https://toppingout.dev/skills/f-inspect.md): Have specialist reviewers check the finished job's changes and report only problems. - [f-night-shift](https://toppingout.dev/skills/f-night-shift.md): Run several jobs one after another without you: a shift boss opens a fresh foreman for each. - [f-plancheck](https://toppingout.dev/skills/f-plancheck.md): Have AI models from other companies read the plan fresh and point out problems. - [f-schedule](https://toppingout.dev/skills/f-schedule.md): Split a big plan into jobs, in order, and draw the project board. - [f-search](https://toppingout.dev/skills/f-search.md): Find Logbook cards about a topic. - [f-setup](https://toppingout.dev/skills/f-setup.md): Install the method into a repo or upgrade it, write the repo profile, and prove the install. - [f-shift](https://toppingout.dev/skills/f-shift.md): Start a session as the foreman: read the role, the rules and the playbook first. - [f-signoff](https://toppingout.dev/skills/f-signoff.md): Close a merged job: check every requirement passed, save the lessons, file the job away. - [f-sketch](https://toppingout.dev/skills/f-sketch.md): Write down a rough idea and its open questions, before anything is decided. - [f-status](https://toppingout.dev/skills/f-status.md): Tell you in plain words where things stand and what's next. - [f-survey](https://toppingout.dev/skills/f-survey.md): Turn a rough request into clear decisions, one round of questions at a time. - [model-council](https://toppingout.dev/skills/model-council.md): Ask a panel of AI models, each answering on its own, to help settle a design choice. - [orca-pi-connect](https://toppingout.dev/skills/orca-pi-connect.md): Link the foreman to a Pi builder in the Orca pane beside it, and know when it finishes. - [pack-review](https://toppingout.dev/skills/pack-review.md): Bundle the goal, the rules and the changes into one brief for an outside reviewer. - [speak](https://toppingout.dev/skills/speak.md): Read one reply out loud as a short spoken summary. - [xreview](https://toppingout.dev/skills/xreview.md): Review another session's work with fresh eyes: list each claim, never rewrite it. ## Optional - [Every skill and agent page in one file](https://toppingout.dev/llms-full.txt): the Markdown twins above, concatenated
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

