sglang-hicache SKILL.md skillA
sglang-hicache from air-gapped/skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add air-gapped/skills/sglang-hicache@git:20261027.42d4297Pins git:20261027.42d4297 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- air-gapped/skills
- Path
.claude/skills/sglang-hicache/SKILL.md- Versions recorded
- 11
- Latest version
- git:20261027.42d4297, committed Oct 27, 2026
- First version recorded
- Sep 15, 2026
- Size
- 30,164 bytes
- Repository stars
- 6
- License
- MIT
SGLang HiCache (hierarchical KV cache) — three-tier prefix cache: GPU HBM (L1) → pinned host DRAM (L2) → distributed L3 (Mooncake / 3FS / NIXL / AIBrix / EIC / SiMM / file / LMCache). Covers `--enable-hierarchical-cache`, all `--hicache-*` flags, write policies, page_first* layouts, prefetch policy (best_effort / wait_complete / timeout), per-rank sizing, MHA / MLA / DSA / Mamba / SWA support matrix (SWA + 3FS hybrid shipped in v0.5.11), runtime attach/detach HTTP admin, and auto-rewrite…
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# SGLang HiCache — hierarchical KV cache Target audience: operators running SGLang on H100/H200/H20/B200-class datacenter GPUs in production, especially long-context agentic workloads (coding agents, RAG, multi-turn chat) with strong prefix locality. ## Why this matters Long-context inference is almost always KV-cache bound, not compute bound. HiCache extends effective KV capacity beyond HBM through pinned host DRAM (L2) and an optional distributed L3 (Mooncake / 3FS / NIXL / AIBrix / EIC / SiMM / file). Reported gains: TTFT –56% / throughput ×2 (Novita Qwen3-Coder-480B + 3FS), TTFT –84% on cache hits (Ant Group DeepSeek-R1-671B + Mooncake) — see [LMSYS blog 2025-09-10](https://www.lmsys.org/blog/2025-09-10-sglang-hicache/). LMSYS's headline "up to 6× / up to 80%" is uncited — trust the deployment-specific numbers. …
Read the whole file at its exact version.
How to install
mdr add air-gapped/skills/sglang-hicache@sha256:75a0b4f8436aa225The line at the top of this page pins the label git:20261027.42d4297, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_vu3e5jjlj4q4io6f)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 11 versions of this file, between Sep 15, 2026 and Oct 27, 2026, about one every 4 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261027.42d4297 latest | 2026-10-27 | 42d4297 | 30,164 B | A | view · diff |
| git:20261027.3796827 | 2026-10-27 | 3796827 | 30,076 B | A | view · diff |
| git:20261027.78212f8 | 2026-10-27 | 78212f8 | 30,076 B | A | view · diff |
| git:20261027.cd7fe1c | 2026-10-27 | cd7fe1c | 29,820 B | A | view · diff |
| git:20261027.d19acd5 | 2026-10-27 | d19acd5 | 29,698 B | A | view · diff |
| git:20260922.5e8f509 | 2026-09-22 | 5e8f509 | 28,175 B | A | view · diff |
| git:20260915.300ba48 | 2026-09-15 | 300ba48 | 28,128 B | A | view · diff |
| git:20260915.5546669 | 2026-09-15 | 5546669 | 27,969 B | A | view · diff |
| git:20260915.802a771 | 2026-09-15 | 802a771 | 27,399 B | A | view · diff |
| git:20260915.b2dd0a6 | 2026-09-15 | b2dd0a6 | 27,355 B | A | view · diff |
| git:20260915.c218485 | 2026-09-15 | c218485 | 25,522 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (30164 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
air-gapped/skills · 6 stars · license MIT · pushed 2026-11-04 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_vu3e5jjlj4q4io6f GET https://markdownregistry.com/api/v1/resolve?ref=air-gapped/skills/sglang-hicache GET https://markdownregistry.com/api/v1/blob/75a0b4f8436aa22573267e2d2f3a3cca24cca3e904ee6621877463d24479f57e
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.