axiom-shell/axiom
A Wayland desktop environment you design on screen. Bars, menus, widgets, lock and login screens for Hyprland, niri and sway.
stack: Python
ARCHETYPE
Skill Collector
Ten-plus skills loaded on demand. A procedural-knowledge library.
GUARDRAILS
3/5
Blocks destructive commands · Pre-tool screening hook · No YOLO mode · details
Quality 50/100 Guardrails v2
Copy this rig
# review before running: this installs third-party code
$ npx degit axiom-shell/axiom/.claude ./rig-axiom # inspect, then merge into .claude/ MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit axiom-shell/axiom/.claude/skills/bar-popout .claude/skills/bar-popout $ npx degit axiom-shell/axiom/.claude/skills/bar-widget .claude/skills/bar-widget $ npx degit axiom-shell/axiom/.claude/skills/compositor-backend .claude/skills/compositor-backend $ npx degit axiom-shell/axiom/.claude/skills/config-section .claude/skills/config-section $ npx degit axiom-shell/axiom/.claude/skills/launcher-command .claude/skills/launcher-command $ npx degit axiom-shell/axiom/.claude/skills/methods-helper .claude/skills/methods-helper $ npx degit axiom-shell/axiom/.claude/skills/migration .claude/skills/migration $ npx degit axiom-shell/axiom/.claude/skills/osd-system .claude/skills/osd-system $ npx degit axiom-shell/axiom/.claude/skills/overlay-module .claude/skills/overlay-module $ npx degit axiom-shell/axiom/.claude/skills/overlay-view .claude/skills/overlay-view $ npx degit axiom-shell/axiom/.claude/skills/review-branch .claude/skills/review-branch $ npx degit axiom-shell/axiom/.claude/skills/review-system .claude/skills/review-system $ npx degit axiom-shell/axiom/.claude/skills/service .claude/skills/service $ npx degit axiom-shell/axiom/.claude/skills/setting .claude/skills/setting $ npx degit axiom-shell/axiom/.claude/skills/settings-card .claude/skills/settings-card $ npx degit axiom-shell/axiom/.claude/skills/theme-integration .claude/skills/theme-integration $ npx degit axiom-shell/axiom/.claude/skills/verify .claude/skills/verify
$ curl -fsSL --create-dirs -o .claude/agents/architect.md https://raw.githubusercontent.com/axiom-shell/axiom/main/.claude/agents/architect.md $ curl -fsSL --create-dirs -o .claude/agents/checker.md https://raw.githubusercontent.com/axiom-shell/axiom/main/.claude/agents/checker.md $ curl -fsSL --create-dirs -o .claude/agents/implementer.md https://raw.githubusercontent.com/axiom-shell/axiom/main/.claude/agents/implementer.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/axiom-shell/axiom/main/.claude/agents/reviewer.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"hooks": {
"PreToolUse": [
{
"matcher": "Bash|Edit|Write",
"hooks": [
{
"type": "command",
"command": "python3 \"$CLAUDE_PROJECT_DIR/.claude/hooks/guard.py\""
}
]
}
]
}
} Skills (17)
bar-popoutbar-widgetcompositor-backendconfig-sectionlauncher-commandmethods-helpermigrationosd-systemoverlay-moduleoverlay-viewreview-branchreview-systemservicesettingsettings-cardtheme-integrationverify
Subagents (4)
| architect model: opus | Designs before code is written and judges what others found. Use to turn a request into an exact spec, to find the cause of an unclear bug, to decide between approaches, or to confirm or reject review |
| checker model: haiku | Runs axiom's checks (scripts/dev/verify.sh, tests, the shell log) and reports only what failed. Use to verify a change without filling the caller's context with check output. |
| implementer model: sonnet | Writes code from a clear spec or a recipe in docs/recipes/, then runs scripts/dev/verify.sh until it passes. Use for any change whose files and shape are already decided. Stops and reports instead of |
| reviewer model: sonnet | Reviews a small slice of a diff through one lens (correctness, rule breaks, dead or repeated code, simplification, stale docs) and returns findings with file:line and a failing case. Read-only. Used b |
Hooks (1)
| event | matcher | runs |
|---|---|---|
| PreToolUse | Bash|Edit|Write | python3 "$CLAUDE_PROJECT_DIR/.claude/hooks/guard.py" |
Permissions
deny (0)
—
ask (0)
—
allow (17)
Bash(scripts/dev/verify.sh)
Bash(scripts/dev/verify.sh *)
Bash(scripts/dev/shot.sh *)
Bash(scripts/log.sh)
Bash(scripts/log.sh *)
Bash(scripts/run_tests.sh)
Bash(scripts/run_tests.sh *)
Bash(scripts/check_all.sh)
Bash(scripts/check_all.sh *)
Bash(scripts/check_qmlformat.sh)
Bash(python3 scripts/check_*.py)
Bash(python3 scripts/check_*.py *)
Bash(python3 scripts/gen_docs.py)
Bash(python3 scripts/gen_docs.py *)
Bash(python3 scripts/check_doc_drift.py)
Bash(python3 scripts/check_doc_drift.py *)
Bash(python3 tests/scripts/test_guard.py)
Similar rigs
Trolzie/claude-loop-systems-demo
A small demo of Markdown-as-system AI dev loops with Claude Code: issues → agents → skills → verification.
Pragmatist 480 tok · · Q15
KirilLukyanenko/comp-ledger
Multi-tenant compensation service written by Claude Code agents from specs (PRPs), checked by holdout scenarios they never see.
Minimalist 1.2k tok · · Q24
yuroka-studio/nightshift-agents
Scheduled Claude Code agents that take one task from a backlog, open a PR with real test evidence, and never merge or deploy on their own.
Fort Knox 1.0k tok · · Q61
homeofe/improvements
Portable, LLM-agnostic framework for multi-model AI workflows: model routing, structured agent handoff (AAHP), and a phased agent pipeline. MIT.
Orchestrator 1.4k tok · · Q24