Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/jim4226/claude-workspace-brainnpx agentmods add commands/jim4226/claude-workspace-brain/brain-archiveWrote 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-archive)<a href="https://agentmods.dev/commands/jim4226/claude-workspace-brain/brain-archive"><img src="https://agentmods.dev/badge/commands/jim4226/claude-workspace-brain/brain-archive/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/jim4226/claude-workspace-brain/brain-archive"><img src="https://agentmods.dev/badge/commands/jim4226/claude-workspace-brain/brain-archive.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00016 | $0.00612 |
| Opus 5 | $0.00008 | $0.00306 |
| Sonnet 5 | $0.00003 | $0.00122 |
| Haiku 4.5 | $0.00002 | $0.00061 |
Grade A, and why
brain-archive 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 9d 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 — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Keep WORKSPACE_BRAIN.md lean by moving aged-out content to a sidecar archive.
The active brain stays small (fast to inject); the archive preserves history
without bloating every session.
Step 1 — survey
Read WORKSPACE_BRAIN.md and identify:
- Completed threads in
ACTIVE THREADSmarked(done),[x], or a check. Anything older than 14 days is a strong archive candidate. - Sessions in
RECENT SESSIONSbeyond the 10 newest entries. - Decisions in
DECISIONS LOGolder than 90 days that aren't load-bearing for current decisions (skim the SYNAPSES section to check for references). - Resolved open questions — these should already be removed when answered; flag any that look stale (>30 days old, no recent activity nearby).
Report a preview to the user: "I propose to move threads, sessions, and decisions to archive. Total: ~ lines, ~ KB."
Step 2 — wait for approval
Do not edit anything yet. Ask the user one of:
- "Apply the migration as proposed?"
- "Adjust the cutoff? (default: 14d threads / 10 sessions / 90d decisions)"
- "Skip - cancel."
Wait for their answer.
Step 3 — execute the migration
If approved:
-
If
WORKSPACE_BRAIN_ARCHIVE.mddoesn't exist, create it with the header:# WORKSPACE BRAIN — Archive > Aged-out entries from WORKSPACE_BRAIN.md. Not auto-injected. > Append-only; oldest at bottom.Use Write for creation. Otherwise use Edit to insert at the top (newest archived first).
-
For each archived entry, prepend to the archive under a date-stamped section:
## Archived 2026-MM-DD, then the entries grouped by their original section name (### From ACTIVE THREADS, etc.). -
Remove the same entries from
WORKSPACE_BRAIN.mdusing Edit.
Step 4 — verify
Run the linter on the trimmed brain:
python .claude/scripts/brain_lint.py WORKSPACE_BRAIN.md
Report the new size + score. The archive operation should improve the size budget axis. Don't proceed if the brain ends up with an empty required section — restore those entries and warn the user.
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.
- 9d ago First seen · 70 lines · 16 tokens per session scan A 5f387bb58799
brain-archive is a command published in the GitHub repository jim4226/claude-workspace-brain (1 stars, last pushed 1mo ago), licensed MIT. It adds 16 tokens to every session and 612 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.