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.
git clone --depth 1 https://github.com/jim4226/claude-workspace-brainWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/jim4226/claude-workspace-brain/brain-init)<a href="https://agentmods.dev/commands/jim4226/claude-workspace-brain/brain-init"><img src="https://agentmods.dev/badge/commands/jim4226/claude-workspace-brain/brain-init.svg" alt="Measured on agentmods" height="20"></a>What 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 5.1 | $0.00014 | $0.00687 |
| Opus 5 | $0.00007 | $0.00344 |
| Sonnet 5 | $0.00003 | $0.00137 |
| Haiku 4.5 | $0.00001 | $0.00069 |
Grade A, and why
brain-init 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 7d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bootstrap a workspace brain for this project. Goal: take the user from "empty template" to "useful brain" in under 60 seconds.
Step 1 — check current state
Read WORKSPACE_BRAIN.md at the project root.
- If it doesn't exist (the installer normally places it at the project root
already), fetch the template from
https://raw.githubusercontent.com/jim4226/claude-workspace-brain/main/template/WORKSPACE_BRAIN.md
and write it to
WORKSPACE_BRAIN.mdat the project root. - If it exists AND has already been customised (any section has non-template
content), stop and report back: "Brain already initialised. Use
/brainto edit a section, or/brain-gradeto audit quality." Do not proceed. - If it exists but is still the unedited template, proceed.
Step 2 — ask the user 6 fast questions
Use AskUserQuestion (if available) or plain prompts. Keep each question to one line. Wait for the user's answer before asking the next.
- Project type? (web app | research / paper | ML training | CLI tool | other)
- One-line goal: what are we building, in one sentence?
- Current sprint focus: what are you trying to ship this week / cycle?
- One in-flight thread: name one piece of work that's currently happening, with status (in-progress / blocked / pending).
- One recent decision + the why: what's a judgment call you made in the last week and why? (Required - decisions without rationale rot fast.)
- One open question: what's still unresolved?
Step 3 — fill the template
Use the Edit tool to populate the brain:
ACTIVE FOCUS→ answer 2 + answer 3ACTIVE THREADS→ one entry from answer 4DECISIONS LOG→ one entry from answer 5, formatted- **YYYY-MM-DD**: <decision>. Reason: <why>.(use today's date)OPEN QUESTIONS→ one entry from answer 6- Set
Last sync:in the header to today's date - Leave
SYNAPSES,KEY NUMBERS,RECENT SESSIONSwith their template placeholders for now - the user can fill them as they accumulate.
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.
- 7d ago First seen · 62 lines · 14 tokens per session scan A bdae69a7393b
brain-init is a command published in the GitHub repository jim4226/claude-workspace-brain (1 stars, last pushed 1mo ago), licensed MIT. It adds 14 tokens to every session and 687 once invoked, about $0.0001 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 commands, from other repositories
load-session
Load a synced session summary for context.
capture
Model-driven lesson capture — read recent work in a role, distill genuine insights, append tagged bullets to pending.md. The real capture path; complements the Stop-hook's cheap regex fast-lane.
role-promote
Autonomously reconcile a role's memory — merge pending + recent learnings into ROLE.md, dedupe and declutter. The auto-promotion step fired when a role is stale (>7 days) with pending activity.
update-memory
Check and update CLAUDE.md memory based on changes to skills, commands, agents, and hooks.
init-workspace-flow-questions
Phase 8 Questions of init-workspace-flow.
export-closedloop-learnings
Exports pending ClosedLoop learnings to global location with deduplication.