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/itsribbz/godspeed/brain-scoregit clone --depth 1 https://github.com/itsribbZ/GodspeedWrote 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/itsribbz/godspeed/brain-score)<a href="https://agentmods.dev/commands/itsribbz/godspeed/brain-score"><img src="https://agentmods.dev/badge/commands/itsribbz/godspeed/brain-score.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 | $0.00027 | $0.00370 |
| Opus 5 | $0.00014 | $0.00185 |
| Sonnet 5 | $0.00005 | $0.00074 |
| Haiku 4.5 | $0.00003 | $0.00037 |
Grade A, and why
brain-score 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 4d 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.
What it actually says
/brain-score — Toke Brain Prompt Classifier
Classify the prompt in $ARGUMENTS against the live routing manifest.
Arguments received: $ARGUMENTS
Execution
Run this exact command via the Bash tool and echo the full output verbatim — no preamble, no rewrite, no summary. Users want the raw classifier signal.
python ${CLAUDE_PLUGIN_ROOT}/automations/brain/brain_cli.py score "$ARGUMENTS"
Output contract
brain score prints in this shape (do not reformat):
Tier: S<N>
Model: <haiku|sonnet|opus> Effort: <low|medium|high>
Score: 0.<NNN>
Reason: score=... | top:<signals> | <escalation-notes>
Signals:
prompt_length 0.NNN #####
code_blocks 0.NNN
file_refs 0.NNN
reasoning 0.NNN
...
Guardrails fired: <list, if any>
Rules
- NEVER ask for confirmation — execute immediately.
- If
$ARGUMENTSis empty, print:usage: /brain-score <prompt text>and stop. - If the Python command errors, show stderr verbatim — never swallow the error.
- This command is read-only. No file writes, no settings edits.
- Do not add interpretation beyond the raw classifier output unless the user explicitly asks for analysis after.
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.
- 4d ago First seen · 48 lines · 27 tokens per session scan A da62af35f30f
brain-score is a command published in the GitHub repository itsribbZ/Godspeed (1 stars, last pushed 2mo ago), licensed MIT. It adds 27 tokens to every session and 370 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
stats
Show model-routing stats - dispatches and real token volume kept off the session model. Args - --days N (window size, default 7), --ago M (shift window M days back), --session (only sessions on that model, e.g. fable).
doctor
Diagnose what is busting Claude Code's DeepSeek prefix cache.
benchmark
Run the Permafrost cache benchmark (offline emulator, no API key).
status
Show Permafrost's live DeepSeek cache hit rate and dollar savings.
wrap
Explain how to launch Claude Code through the Permafrost proxy.
init
Install the formatters this repository needs, with every command visible before it runs.