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 commands/brainos-hq/brain-os/startgit clone --depth 1 https://github.com/brainOS-HQ/brain-osWrote 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/brainos-hq/brain-os/start)<a href="https://agentmods.dev/commands/brainos-hq/brain-os/start"><img src="https://agentmods.dev/badge/commands/brainos-hq/brain-os/start.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 | $0.00000 | $0.00891 |
| Opus 5 | $0.00000 | $0.00445 |
| Sonnet 5 | $0.00000 | $0.00178 |
| Haiku 4.5 | $0.00000 | $0.00089 |
Grade A, and why
start 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 3d 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 — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Brain OS : Session Start
"What accumulated? What's blocking? Where do I begin?"
Run this at the start of every work session. It surfaces everything that accumulated since last time — including anything an auto-wrap saved without review — and gives you one clear place to begin.
REQUIRED FIRST READ
Before any tool call, read ~/.claude/brain-os/PROTOCOL.md.
Input
Arguments: $ARGUMENTS (optional project name to scope)
Step 0: Reconcile unconfirmed auto-wraps & checkpoints
Before anything else, check for state the last session saved without review:
.brain/sessions/autowrap-*.json (auto-wraps from wrap_auto)
.brain/sessions/checkpoints/*-*.json (raw captures from the PreCompact / SessionEnd hooks)
Look for files where "confirmed": false.
- Auto-wraps (
autowrap-*.json): theappliedfields were already written (low-risk — next move, open questions, evidence). Thepending_reviewitems were held back for you. For each high-risk item (status, mode, blocked, decisions), ask one short yes/no, then apply confirmed ones via the normal tools and mark the record"confirmed": true. - Raw checkpoints: summarize what was captured (
last_user_goals,open_questions,last_assistant_summary) and offer to fold it into a proper wrap. Mark"confirmed"or"discarded"after.
Surface these under "RECOVERED FROM LAST SESSION" in the output. If none exist, skip silently.
Tool sequence
mcp__brain-os__context_resolve(user_message=$ARGUMENTS)— resolve scope if project namedmcp__brain-os__entity_read(entity_id)— current state, open questions, blockersmcp__brain-os__plan_read(entity_id)— active plan step and progressmcp__brain-os__decision_review(entity_id, limit=3)— overdue decisionsmcp__brain-os__focus_get(entity_id)— priority recommendation
Run all 5. Combine into a single session brief.
Output format
SESSION START — [date]
═══════════════════════════════════════
[Project / Global]
RECOVERED FROM LAST SESSION
✓ Auto-saved: [what was already applied]
? Needs your call: [staged high-risk item — confirm/discard]
SINCE LAST SESSION
Plan: [X/N steps done] — active: [current step]
[If anything changed: "Last updated [N] days ago"]
NEEDS ATTENTION NOW
🔴 [blocker or overdue decision — act before starting]
🟡 [stale question or aging open item]
FOCUS
→ [One concrete thing to do right now, from focus_get]
PIPELINE
→ Active: [current plan step]
→ Next: [next step]
OPEN QUESTIONS (oldest first, unresolved)
? [question — if older than 7 days, mark as "needs resolution"]
? [question]
OVERDUE DECISIONS
- "[decision]" — due [date] (use /reconcile)
═══════════════════════════════════════
Start with: [exact first action]
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.
- 3d ago First seen · 97 lines · 0 tokens per session scan A 8216d9d612a1
start is a command published in the GitHub repository brainOS-HQ/brain-os (6 stars, last pushed 10d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 891 tokens. 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
rp-reminder-cli
Continue your current workflow using rp-cli instead of built-in alternatives.
summarize-channel
Summarize recent activity in a Slack channel.
tree-ring-recall
Recall durable Tree Ring Memory context before starting or resuming work.
session
Complete reference for all session-related commands — start, end, list, and compact.
work-resume
중단된 Active Work를 재개한다. drift 확인 후 남은 계획을 제안한다.
setup-gogcli
Command "setup-gogcli" from minicoohei/ai-agent-camp, covering gogcli (google workspace cli) セットアップ, step 0: セットアップ進捗の確認, このセッションでやること, 準備チェック and step 1: gogcli のインストール.