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 agentmods add skills/axera-tech/magnetar/acquirenpx skills add AXERA-TECH/Magnetar --skill acquiregit clone --depth 1 https://github.com/AXERA-TECH/MagnetarWrote 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/axera-tech/magnetar/acquire)<a href="https://agentmods.dev/skills/axera-tech/magnetar/acquire"><img src="https://agentmods.dev/badge/skills/axera-tech/magnetar/acquire.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00025 | $0.00803 |
| Opus 5 | $0.00013 | $0.00402 |
| Sonnet 5 | $0.00005 | $0.00161 |
| Haiku 4.5 | $0.00003 | $0.00080 |
Grade A, and why
acquire scanned grade A with 1 finding 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 4d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
如 tokenizer.json 可直接 `curl/wget $HF_ENDPOINT/...` 单线获取)。 What it actually says
ACQUIRE
执行
magnetar.stages.acquire.run(task_dir, source)
下载镜像默认:ModelScope 优先(国内 CDN);HuggingFace 走 HF_ENDPOINT(默认 hf-mirror);
Git URL 克隆经 GH_PROXY(默认 gh-proxy);uv/pip 用 PIP_INDEX_URL(默认阿里云)。
全部可在 .magnetarrc / 环境变量覆盖,置空字符串禁用。
SOURCE 为 HF repo 时,先 magnetar.net_util.modelscope_available("<org>/<name>")
探测 ModelScope 是否有同名仓库,有则 modelscope download --model <org>/<name> 获取,
没有才回退 HuggingFace(hf-mirror)。
HF 大文件(权重/大附件)回退时用 hf-mirror 的 hfd 工具多线程下载:
scripts/download_hf.sh <org>/<name> --local-dir origin/<name> -x 8
(自动缓存 ~/.cache/magnetar/hfd.sh,HF_ENDPOINT 默认 hf-mirror;小文件
如 tokenizer.json 可直接 curl/wget $HF_ENDPOINT/... 单线获取)。
拿到模型后,记录运行流程:调用 magnetar.stages.acquire.write_model_flow(task_dir, flow)
写入 origin/model_flow.json,字段见函数 docstring。至少包含:
example_input:真实样本路径(保证 SDK 用与验证一致的数据)preprocess_code/postprocess_code:预处理/后处理函数体(SDK 原样嵌入)verified: true:表示该流程已实际跑通模型
SDK-GEN 阶段必须读取此文件生成 SDK,确保与 ACQUIRE 验证过的运行流程一致; 缺失或示例样本不存在时 SDK 生成会报错。
LLM/自回归检测(model_route gate 的输入)
拿到模型后用 magnetar.stages.llm.classify(origin, source, model_name) 判定并把结果
写入 cache/acquire/manifest.json 的 route_hint:
{"llm": true|false, "reason": "...", "hybrid": true|false}
检测信号(命中任一即 route=llm):
config.json的architectures/model_type(含嵌套text_config)为 causal LM, 如Qwen2ForCausalLM、LlamaForCausalLM、MiniCPMForCausalLM- README.md 的
pipeline_tag: text-generation model_flow.json的 task 为 text_generation / causal_lm / chat- SOURCE/MODEL_NAME 含已知 LLM 名称;MOSS、NeuTTS、VALL-E、Audio8 等为 hybrid
(整体 TTS 但含 LLM/AR 骨干,需拆分,
hybrid: true)
验证
origin/下有模型文件或source.txtACQUIRE_REPORT.md已生成model_flow.json已记录运行流程(缺失时 SDK-GEN 只能生成直通预处理/后处理,需 Agent 注意)route_hint已写入 manifest.json(供 model_route gate 复用)
STOP
- SOURCE 无效(本地路径不存在、URL 不可达、Git clone 失败)
- 需要私有凭据(HF_TOKEN 缺失)
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.
- 4d ago First seen · 54 lines · 25 tokens per session scan A d54970b4df48
acquire is a skill published in the GitHub repository AXERA-TECH/Magnetar (22 stars, last pushed yesterday), licensed MIT. It adds 25 tokens to every session and 803 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…