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/yucai0302/memory-loop/compressgit clone --depth 1 https://github.com/yucai0302/memory-loopWhat 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 | $0.00029 | $0.00105 |
| Opus 5 | $0.00015 | $0.00053 |
| Sonnet 5 | $0.00006 | $0.00021 |
| Haiku 4.5 | $0.00003 | $0.00011 |
Grade A, and why
compress 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.
The source is not reproduced here
No licence file
A repository with no LICENSE is all rights reserved by default, so the body is not copied here. The metadata, the measurements and the link are.
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 · 12 lines · 29 tokens per session scan A 34d7fa8b35d4
compress is a command published in the GitHub repository yucai0302/memory-loop (30 stars, last pushed 2mo ago), with no licence file. It adds 29 tokens to every session and 105 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 2026-08-30.
Other commands, from other repositories
tokenless-stats
Show Tokenless compression statistics.
evolve-lite-subscribe
Add a shared guidelines repo (read-scope subscription or write-scope publish target) to the unified repos list.
cb-search
코드브레인 BM25 검색 — stale 자동 refresh.
graphify
Turn your vault into a clustered knowledge graph with HTML and JSON outputs.
/reddit drives the visible Chrome CDP session for Reddit. It does not use the Reddit API and never launches a headless browser. It acts as whatever Reddit account is logged into the visible session.
scorecard
Run a repeatable weekly health check. Score Brain OS out of 100. Compare week-over-week.