ask
Answer a question from this project's own knowledge — retrieved, read and cited, never from memory.
Take a folder that already holds research — papers, notes, code, half-written drafts — into MAGI, without breaking what is already there.
$ npx -y skills add Misaka16384/Wikify --skill adopt --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/adoptContext preview
The summary Claude sees to decide when to auto-load this skill.
Take a folder that already holds research — papers, notes, code, half-written drafts — into MAGI, without breaking what is already there.
name: adopt description: "Take a folder that already holds research — papers, notes, code, half-written drafts — into MAGI, without breaking what is already there." commands: adopt: "Bring an existing research folder into MAGI." origin: magi
A folder with work already in it and no `threads/`. A project MAGI itself made is `magi migrate`, not this.
1. `magi adopt survey .` — read the folder first. It also pulls every arXiv id and DOI out of the prose, which is where most of the sources already are. 2. Collect what this conversation says about the work: papers named, claims made. Nothing there is a normal result — say "0 from the conversation". 3. `magi init` in place. It only adds; it touches nothing already there. 4. Write `{"moves": [{"from": "x", "to": "drafts/x"}]}` — others' material to `raw/`, the human's own working out to `drafts/`, code and data left alone. Shape it to MAGI: the references between the files are repaired for you. 5. Show the human the plan, then `magi adopt apply plan.json --dry-run`, then for real. `magi adopt undo` puts the files *and* the edited text back. 6. `magi ingest url <the ids>`, `magi ingest batch-run`, then `ingest`. 7. What matters: open what the material claims. Worked out already, it is a **finding**, no bet to ask: `magi thread new <slug> --kind proposition --title '<claim>' --purpose '<why now>' --found <date> --derivation <draft>`. Only an open conjecture takes `--bet unknown`. A plan index and a status board are lines and propositions already. 8. `magi index`, then `magi next`. Still "no propositions" means the folder was tidied, not adopted.
hand-edited, so a draft filed there can never be worked on again.
project this is — that one is not yours to decide.
MAGI 是一个 agent-native 的科研工作环境:人是驾驶员,LLM agent 是机体,确定性的 magi CLI 是拘束具——三者同步率越高,科研越快。它把学术论文(PDF/LaTeX)摄入、编译为 Obsidian 兼容的概念卡片项目,并用三核架构管理完整的科研状态: 进入任意项目先跑 magi sync——它输出同步率、三核状态和逐条可执行的修复提示。magi radar 是文献雷达:定时发现相关新论文,并侦察"知识上应引用我方论文却未引用"的候选。同一份 skills 通吃 Claude
Repo: Misaka16384/Wikify
Answer a question from this project's own knowledge — retrieved, read and cited, never from memory.
Compile raw sources into dense, interlinked reference and concept cards, and mine the concepts a compiled card left implicit.
Turn whatever the human has — a PDF, a link, a DOI, a screenshot of a citation — into raw/ sources this project can compile.
Entry point for a MAGI research project: run magi next, do what it says, and call the skill it names.
Triage a literature-radar digest: judge each candidate against what this project is actually doing, and queue only what earns it.