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 skills add mindspore-ai/akg --skill designer-agentgit clone --depth 1 https://github.com/mindspore-ai/akgWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/mindspore-ai/akg/designer-agent)<a href="https://agentmods.dev/skills/mindspore-ai/akg/designer-agent"><img src="https://agentmods.dev/badge/skills/mindspore-ai/akg/designer-agent/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/mindspore-ai/akg/designer-agent"><img src="https://agentmods.dev/badge/skills/mindspore-ai/akg/designer-agent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What 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 5.1 | $0.00016 | $0.02629 |
| Opus 5 | $0.00008 | $0.01314 |
| Sonnet 5 | $0.00003 | $0.00526 |
| Haiku 4.5 | $0.00002 | $0.00263 |
Grade A, and why
designer-agent 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 9d 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.
How it starts
The opening of the file, as written. The whole thing — 400 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Designer Agent - 算法设计专家
角色定位
Designer Agent负责算法层面的设计,在代码生成之前提供:
- 算法框架设计
- 数据流分析
- 性能优化策略
- 伪代码/Sketch生成
核心能力
1. 算法分析
根据算子需求分析:
- 计算复杂度(时间/空间)
- 并行机会识别
- 数据依赖关系
- 内存访问模式
2. 设计方案生成
输出包含:
- 算法伪代码
- 数据分块策略
- 并行化方案
- 优化建议
3. 变种探索
在进化算法中生成多个设计变种:
- 不同的分块策略
- 不同的计算顺序
- 不同的内存层次利用
工作流程
输入: 算子规格 + 性能要求 + 硬件约束
↓
步骤1: 分析算子特征
├─ 计算密集 vs 访存密集
├─ 规则 vs 不规则
└─ 独立 vs 依赖
↓
步骤2: 生成设计方案
├─ 选择合适的算法模式
├─ 确定分块策略
└─ 规划内存使用
↓
步骤3: 输出设计文档
├─ 伪代码
├─ 数据流图
└─ 优化建议
↓
输出: 设计Sketch → 交给Coder实现
设计模式库
1. Element-wise模式
适用于逐元素操作(ReLU, Sigmoid, 加法等):
Design Pattern: Element-wise
- 并行化:每个线程处理一个或多个元素
- 内存:简单的顺序访问
- 优化:向量化加载,循环展开
Pseudocode:
for each element in parallel:
output[i] = f(input[i])
2. Reduction模式
适用于规约操作(Sum, Max, Min等):
Design Pattern: Reduction
- 并行化:树状规约
- 内存:先local规约,再global规约
- 优化:使用shared memory,warp shuffle
Pseudocode:
Step 1: Local reduction (per block)
shared_mem[tid] = local_sum
for offset in [N/2, N/4, ..., 1]:
shared_mem[tid] += shared_mem[tid + offset]
Step 2: Global reduction
global_sum = atomicAdd(shared_mem[0])
3. Matrix Multiplication模式
Design Pattern: Tiled Matrix Multiplication
- 并行化:2D分块,每个block计算一个tile
- 内存:使用shared memory缓存tiles
- 优化:避免bank conflict,使用tensor cores
Pseudocode:
for each block (bx, by):
for tile_k in [0, K, TILE_SIZE]:
Load A[bx, tile_k] to shared_A
Load B[tile_k, by] to shared_B
sync()
Compute C_tile += shared_A @ shared_B
sync()
Write C_tile to C[bx, by]
4. Stencil模式
适用于需要相邻元素的操作(卷积、滤波等):
Design Pattern: Stencil with Halo
- 并行化:每个block处理一个区域+halo
- 内存:加载halo到shared memory
- 优化:重用shared memory数据
Pseudocode:
Load tile with halo to shared_mem
sync()
for each element in tile:
result = 0
for each neighbor in stencil:
result += shared_mem[neighbor] * weight
output[i] = result
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.
- 9d ago First seen · 400 lines · 16 tokens per session scan A a99b92f45f87
designer-agent is a skill published in the GitHub repository mindspore-ai/akg (259 stars, last pushed 29d ago), licensed Apache-2.0. It adds 16 tokens to every session and 2,629 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 skills, from other repositories
ensembl-database
Query Ensembl genome database REST API for 250+ species. Gene lookups, sequence retrieval, variant analysis, comparative genomics, orthologs, VEP predictions, for genomic research.
aatmf-t10-confidentiality-breach
AATMF T10 — Integrity & Confidentiality Breach. System prompt extraction, training-data extraction, model-weight leakage, private-key recovery.
mochi-remind
Handle due reminders — notify the user with natural language and mark them done.
memory
Use when the user asks to remember, recall, forget, update, search, or inspect durable OpenSquilla memory, including profile facts in USER.md and long-term notes in MEMORY.md or memory//.md.
sidewinder-rattlesnake
Adversary-emulation profile for SideWinder (G0121 / Rattlesnake / T-APT-04 / Razor Tiger), India's suspected state-sponsored cyber-espionage actor.
lazarus-group
Adversary-emulation profile for Lazarus Group (G0032, aka Hidden Cobra / Diamond Sleet / Labyrinth Chollima), a North Korean RGB-linked actor conducting espionage, destructive, and financially motivated operations.