Studio
Load before ANY design work (canvas or code, any project). The single front door to the design practice: the house law (rules.md grades, invariants, motion defaults), the playbook (playbook.md, what to do per situation), the inventory (inventory.md, every tool/skill/mcp and its verdict), and doctrine/ (intake, scaffold, package format), one self-contained package that travels to every project. En…
Install
$ agentstack add skill-tommylower-cortex-studio ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
About
studio
the law lives HERE, inside this skill folder — one package, every project (law self-contained 2026-07-05, renamed studio law -> studio and made the front door 2026-07-06, tommy's calls). nothing to chase, nothing installed per project.
read first, every time
paths relative to this skill's base directory:
rules.md— grades, invariants, defaults (including motion numbers)playbook.md— the path through a session: fresh project, existing
project, component work, shipping, stuck
doctrine/component-intake.md— how ideas and references become
components
doctrine/codebase-scaffold.md— only when touching codebase structure
or starting a project
doctrine/design-system-package.md— only when emitting or inheriting
a design-system package
inventory.md— when reaching for a tool or supplier, check its
verdict here first
deposits (one-line experiments earned through pain) land in rules.md here in this folder. the learning journal (friction log + study queue) lives at ~/Developer/code/arc/ (see playbook.md, "when stuck or annoyed").
the loop to enforce
- operator intent first. do not open any reference before the
operator states what the thing should do and feel like.
- references are anatomy-only. shadcn / SDS / base ui / mobbin supply
slots, variant axes, state lists, behavior floors. litmus-test anything crossing over: delete it — breaks behavior or api = structure, take it; changes look/feel only = skin, ours.
- deliverables are state graphs, not happy frames. every state
(default / hover / focus / active / disabled / loading / empty / error) with transitions, motion numbers written in the margins. defaults from rules.md.
- second-use gate. nothing is componentized until it recurs.
- suppliers never decide. cortex skills are catalogs; the law wins
every conflict. motion numbers never come from a skill's defaults.
- no parallel records. design decisions live in the artifact only —
never transcribe them into docs by hand; specs are derived from artifacts. if a mistake costs real time, write a one-line experiment in rules.md at that moment. no close ritual. most sessions deposit nothing.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: tommylower
- Source: tommylower/cortex
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.