setup
Skill5 starsChanged 52 days agoMITView on GitHub
Finish setting up the Katharsis output style after the plugin is installed. Runs the setup script, which checks the Claude Code version, grants the permissions the style's routing step and diagram rendering need, names the two output styles, and reports the diagram and video tools the rich richness level can use, then hands the user the /config step. Use when the user asks to set up, configure, or finish installing Katharsis, or when a session starts with the line "Katharsis is installed but not set up".
- Installs packages
192 other projects use a version of this.
Adopted by
Most recently taken up by
- openstory-so/openstorytheir own version, first seen Nov 16, 2026
- furkandvrc/conductortheir own version, first seen Nov 16, 2026
- futorio/red-greentheir own version, first seen Nov 16, 2026
- JCodesMore/jobs-for-ai-agentstheir own version, first seen Nov 16, 2026
- finolabs/finolabs-vibecodetheir own version, first seen Nov 16, 2026
- vincentventalon/claude-code-video-editing-skilltheir own version, first seen Nov 15, 2026
Used with
Projects with it also use add-atomic-chat-tool, add-dashboard and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
MicrosoftSkilladd-atomic-chat-tool
Add Atomic Chat MCP server so the container agent can call local models served by the Atomic Chat desktop app via its OpenAI-compatible API.
Also in 8 projects that use this
MicrosoftSkilladd-dashboard
Add a monitoring dashboard to NanoClaw. Installs @nanoco/nanoclaw-dashboard and a pusher that sends periodic JSON snapshots.
Also in 8 projects that use this
- +10 more from microsoft
Other versions192 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 192
- anthropics/knowledge-work-plugins★ 36ksame name
- garrytan/gbrain★ 33ksame name
- gavrielc/nanoclaw★ 31ksame name
- nanocoai/nanoclaw★ 31ksame name
- qwibitai/nanoclaw★ 31ksame name
- Q00/ouroboros★ 6.4ksame name
- pytorch/executorch★ 5.2ksame name
- aws/agent-toolkit-for-aws★ 3.5ksame name
- agenticnotetaking/arscontexta★ 3.5ksame name
- spotify/portal-ai-plugins★ 2.8ksame name
182 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository6 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .claude/CLAUDE.mdCLAUDE.md
- commitSkill
- prSkill
- review-replySkill
- rules-checkSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 1 is distinctive, and 2 are very specific to this project.
0
Common
Most of the 169 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 18
If that path does not exist, the SessionStart hook could not make the link, and its message at session start says why: usually a …
- Very specific: line 44
When the script exits 0, its last lines report which diagram renderers (d2, mmdc) and video tools (vhs, ffmpeg, remotion) are on PATH, with …
- Distinctive: line 61
If the Bash call itself is denied, hand the user the command to run in bash mode, where it runs in their shell with no permission prompt:
“Common” is measured: kept by at least 6 in 10 of the 169 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Katharsis setup
Installing the plugin puts the style on disk. Two steps remain, and this skill walks the user
through both. Never edit a settings file by hand: scripts/setup.sh is the single writer, so
running the skill and running the script from a terminal cannot leave different results.
1. Run the setup script
~/.claude/katharsis/scripts/setup.shIf that path does not exist, the SessionStart hook could not make the link, and its
message at session start says why: usually a ~/.claude/katharsis directory left by
Katharsis 0.2.x. Relay that message, ask the user to remove the directory and restart
Claude Code, and stop here.
The script adds two entries to permissions.allow in ~/.claude/settings.json:
Bash(~/.claude/katharsis/scripts/katharsis-exchange-style.sh:*)
Bash(~/.claude/katharsis/scripts/katharsis-artifact.sh:*)The style has you run the first script once per turn, and the second to render a diagram or
reserve and open a video at the rich richness level, and to write a question page the user asks
for at any level. Without an entry, every session prompts for that script on first use.
The script is idempotent and says when an entry is already there.
2. Relay the output
Show the script's output as it printed. When it exits 4, Claude Code is older than 2.1.287, the
oldest version Katharsis supports.
Relay the fix the script printed and stop, because setup does not finish below that version. The
permission was granted anyway, so the rerun only has to confirm the fix. When it reports that the settings file is not valid
JSON, stop and give the user the entries to add by hand, quoted from the output. When it fails
on a missing python3, say so and stop; the Stop hooks need it.
When the script exits 0, its last lines report which diagram renderers (d2, mmdc) and video
tools (vhs, ffmpeg, remotion) are on PATH, with an install command for each one missing; an
exit 3 or 4 prints no report. The script installs nothing. Relay the report as printed, and when a
tool is missing, offer its install command: the text inside the backticks, never the note after
them. A line that names a missing installer instead of a command has nothing to run, and neither
does Remotion's not needed line. Run an install command only after the user says yes to that
command; a yes to one is not a yes to the others. After a yes, run a command that does not contain
sudo yourself. Hand a command that contains sudo to the user to run in a terminal outside Claude
Code, never as a bash-mode ! command, because sudo's password prompt needs a terminal. Hand over
the same way a command that fails with a permission error, such as EACCES, "a terminal is
required", or "permission denied", as an installer such as npm install -g can when its install
location belongs to root, and do not retry it. Remotion counts only as the Katharsis Remotion
project in ~/.claude/katharsis-data/remotion/, which its install command creates with its packages
and browser and marks ready last; a remotion command on PATH does not count. Nothing needs a tool:
without a video tool, rich uses Claude Motion when the session offers it and the option is on,
otherwise a rendered diagram, otherwise an inline sketch.
If the Bash call itself is denied, hand the user the command to run in bash mode, where it runs in their shell with no permission prompt:
! ~/.claude/katharsis/scripts/setup.sh3. Hand over the /config step
Picking the style is the user's own action; nothing in the plugin writes outputStyle. Tell
them to open /config, choose Output style, and pick one of the two:
katharsis:Katharsis, the style alone.katharsis:Katharsis coding, the same style with Claude Code's built-in software-engineering instructions kept.
/config saves the choice to .claude/settings.local.json in the current project. The style
takes effect on the next turn.
In one line, also name the Autonomy level row in /config: it starts at guided, where pushes,
PRs, and messages to colleagues are the user's call, and the drawer suggests standard once their
answers show they usually take the model's recommendation.
4. Offer the rules check
Last, offer /katharsis:rules-check, which finds rules in the project's CLAUDE.md, AGENTS.md,
and other instruction files that repeat or contradict the style and suggests an edit for each.
It changes no file without the user's yes. Offer it in one line and run it only if they accept.
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.

