dsh-token-skills
Cut token cost in 4 layers without cutting model intelligence: handoff, rtk, headroom, caveman token-saving skills — each also available as a standalone installable plugin.
四层省 Token，不牺牲模型智能：handoff、rtk、headroom、caveman 四个省 token 的 skill，每个也是独立可装插件。
How to install
dsh plugin add github:WODE25500/dsh-token-skills About
dsh-token-skills **Cut token cost in 4 layers — without cutting model intelligence.** A DeepSeek Harness preset of four complementary token-saving skills. Each one attacks a different stage of the token pipeline, so they *stack*: **handoff** the context *→ prune* the input *→ compact* the context *→ terse* the output Stop paying twice for tokens you already have. mermaid flowchart LR A[Tokenizer] --> B[① Handoff<br>session] B --> C[② RTK<br>input] C --> D[③ Headroom<br>context] D --> E[④ Caveman<br>output] E --> F[<b>fewer tokens</b>] **Why: every token is money.** A long agent session re-send…
Recommendation signals
Meta
- License
- MIT
- Language
- JavaScript
- GitHub stars
- 1
- mo. downloads
- –
- Last push
- 2026-09-04
- Created
- 2026-08-30
Links
Basic safety check
- Findings
- curated 收录但无 npm 包/安装命令
- Sources
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin
Related plugins
superpowers
obra/superpowers
An agentic skills framework & software development methodology that works.
dev-browser
SawyerHood/dev-browser
A Claude Skill to give your agent the ability to use a web browser
ruflo
ruvnet/claude-flow
🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated