ovg-projectovg-project/kvcached/AGENTS.md

kvcached

AGENTS.md2k starsChanged 33 days agoApache-2.0View on GitHub

Guidance for potential contributors using an AI coding assistant on this repo. Yes, you can use AI to draft a PR description, an issue, or a review comment. BUT! You MUST read it yourself and cut it down before you post it. Unedited AI text is very easy to spot and expensive to review: it is long and complex, and it asserts things nobody checked.

Oct 28, 2025Model Cold-Staring / Sleep - WakeUp / Offloading （模型切换/冷启动/超配） — pacoxu/AI-Infra1 reaction
Oct 18, 2026engine: Flash-Next books the whole KV, hot set stays capped — nibor1896/crow-nest0 reactions · 1 comment
Sep 25, 2026KVcached? (DO NOT DO YET) — wavect/semaprax0 reactions
Sep 24, 2026docs(non-cc): screenshot-lead research batch 11 (plan 0008 row 20) — qte77/ai-agents-research0 reactions
Aug 14, 2026【开源自荐】kvcached:让多个大模型弹性共享一张 GPU 的显存 — ruanyf/weekly0 reactions

10 more on GitHub

Jul 10, 2026feat(radixkv): one signed retention dial for the prefix cache — <0 keep-all / 0 keep-none / >0 evict-to-N — anthony-chaudhary/fak0 reactions
Jul 10, 2026feat(preflight): floor the shrinking dimension to a pending-contraction target in EvaluatePreflight — anthony-chaudhary/fak0 reactions
Jul 10, 2026feat(observability): reclaimable-reserved band + pressure color in fak's live meters (reuse ablateBarCells + ctxvalue 50/80) — anthony-chaudhary/fak0 reactions · 1 comment
Jul 10, 2026feat(dispatch): externally-written live concurrency setpoint polled each tick — grow-now, shrink-on-drain — anthony-chaudhary/fak0 reactions
Jul 10, 2026feat(microagent): two-watermark warm band on hibernation ResidentCap — warm-park before decommit + warm-refill below low-water — anthony-chaudhary/fak0 reactions
Jul 10, 2026feat(microagent): per-key single-flight on HibernationStore.Wake — coalesce a wake stampede on one hibernated id — anthony-chaudhary/fak0 reactions · 1 comment
Jun 21, 2026maintenance: weekly update 2026-06-21 — pleasedodisturb/awesome-llm-token-optimization0 reactions · 1 comment
Jun 9, 2026[RFC]: Control-plane orchestration of vLLM sleep mode and kvcached for warm standby and multi-model GPU pooling — vllm-project/aibrix0 reactions · 9 comments
Nov 15, 2025Project Request: kvcached — InftyAI/Awesome-LLMOps0 reactions · 1 comment
Oct 28, 2025Add model lifecycle management documentation: cold-start, sleep mode, and offloading — pacoxu/AI-Infra0 reactions · 2 comments

In this repositoryOne other file it gives its agents.
- CLAUDE.mdCLAUDE.md

The file

Create a free account to mount

Files

AGENTS.md

Working on kvcached

Guidance for potential contributors using an AI coding assistant on this repo.

Drafting with AI is fine. Pasting the draft is not.

Yes, you can use AI to draft a PR description, an issue, or a review comment. BUT! You MUST read it yourself and cut it down before you post it. Unedited AI text is very easy to spot and expensive to review: it is long and complex, and it asserts things nobody checked.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free