SKILLEMALL.ai
GitHub ↗ RU Sign in

BB compute-mamba-ratio

Compute the optimal --mamba-full-memory-ratio (or --max-mamba-cache-size pin) for a hybrid attention + linear-attention (Mamba / GDN / KDA) model's two serving memory pools, from the workload and serving config. Use when a user asks what ratio to set, why concurrency is clamped, or how to size the state vs KV pools for a hybrid model.

sgl-project/sglang Agent Skills author: sgl-project Apache-2.0 1 file body ≈ 2 849 tokens Open the sourcegithub.com↗ analyzed 5 d ago

Compute the optimal --mamba-full-memory-ratio (or --max-mamba-cache-size pin) for a hybrid attention + linear-attention (Mamba / GDN / KDA) model's two…

As a process B 68/100 · Nearly there — weak spots: result and completion, when it triggers, running it twice

ProcedureSoftware developmentInfrastructuretype and topics are labelled automatically from the skill text
Runs in: Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) OpenClaw Goose OpenCode
JSON
Technical rating
B
94/100
safety, quality, tests
Safety 60%
100
Quality 40%
84
Run on models
none yet
Process rating
B
68/100
Nearly there
Result and completion w 14
0
When it triggers w 12
20
Running it twice w 4
30
the three weakest of ten parameters · all ten

How to improve

    For the model run — optional
    • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
    • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

    Guard findings · 0

    ✓ No critical or high findings

    Files scanned: 1. Evidence is masked. Grey chips explain why severity was lowered.

    Against the Agent Skills spec

    ✓ No remarks against the Agent Skills spec

    Process rating: all ten parameters 68/100

    • 0Result and completion. Does not say what the result is
    • 20When it triggers. No condition that starts the skill
    • 30Running it twice. 4 mutating operations with no state check
    • 70Inputs and preconditions. Inputs and preconditions are listed
    • 85Steps. 31 steps, 1 vague phrases
    • 100Tools and files. No external tools needed
    • 100Failures and branches. 1 branches, has a failure section
    • 100Consistency. Name and required fields are in place
    • 100Execution cost. Instruction body is 2849 tokens
    • 100Progress reporting. Reports progress

    Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

    Quality signals

    • +5Description has no quoted example phrases that should trigger the skill
    • +4Description does not say when NOT to use the skill (false activations)
    • +3Output format is not stated: the model decides each time
    • +1No license
    • +2Single-language instructions
    • +3Description length 336: enough signal without eating the budget
    • +4Structure: 9 headings
    • +3Step-by-step instructions: 31 items
    • +4Has examples (3 code blocks)

    Quality base 70; lint remarks subtract, signals add up to 100. Result: 84.