Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/jbdamask/mcbrain/mcbrain-setupnpx skills add jbdamask/McBrain --skill mcbrain-setupgit clone --depth 1 https://github.com/jbdamask/McBrainWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00405 | $0.13265 |
| Opus 5 | $0.00202 | $0.06633 |
| Sonnet 5 | $0.00081 | $0.02653 |
| Haiku 4.5 | $0.00040 | $0.01327 |
Grade A, and why
mcbrain-setup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 1,122 lines — stays where its author put it; the contents beside it link to each section on GitHub.
McBrain Setup
Sets up McBrain — a personal LLM-maintained knowledge base — end-to-end for Claude Desktop (Cowork) + Obsidian. Pattern is Karpathy's LLM Wiki.
The idea in one sentence: instead of re-deriving knowledge from raw sources every session, Claude builds and maintains a persistent markdown wiki that compounds over time. Obsidian is the IDE; Claude is the programmer; McBrain is the codebase.
⛔ STOP — sandbox self-awareness (read this first)
You (the model running this SKILL) are inside Cowork's Linux sandbox. Your Bash tool sees the sandbox, NOT the user's host machine (Mac or Windows). This is the single biggest source of broken setups, so be explicit with yourself before you run anything:
What the sandbox CANNOT tell you
You cannot detect what's installed on the user's host. Period. The sandbox is a fresh Linux box; what's there has no relationship to what's on the user's Mac or PC.
Forbidden Bash commands in this SKILL
Never run any of these from the Bash tool — the result will mislead you and you will tell the user something false (e.g. "gh isn't installed" when it actually is, just not in the sandbox):
python3 --version which gh gh --version
python --version which git git --version
which python / python3 which node node --version
which brew brew --version xcode-select -p
which winget winget --version which rg / rg --version
which pip / pip3 pip --version command -v <anything>
If you catch yourself thinking "let me just check if X is installed" — STOP. The check is meaningless. Instead:
- For tools the user runs themselves (gh, git, brew, winget, python,
node, ripgrep): ask the user to run the version command in their
own Terminal / PowerShell and paste the output. Or just instruct
them to install it (with the right command for their
OS_TYPE) and trust that they did. If they're wrong, the failure surfaces later with a clear error — they'll fix it and retry. That's a better UX than you wrongly reporting "X isn't installed" right now.
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 1,122 lines · 405 tokens per session scan A 2d26a8a38eed
mcbrain-setup is a skill published in the GitHub repository jbdamask/McBrain (2 stars, last pushed 2mo ago), licensed MIT. It adds 405 tokens to every session and 13,265 once invoked, about $0.0020 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…