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/gyf0311/lorekit/wiki-ingestnpx skills add GYF0311/lorekit --skill wiki-ingestgit clone --depth 1 https://github.com/GYF0311/lorekitWrote 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/gyf0311/lorekit/wiki-ingest)<a href="https://agentmods.dev/skills/gyf0311/lorekit/wiki-ingest"><img src="https://agentmods.dev/badge/skills/gyf0311/lorekit/wiki-ingest.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 | $0.00062 | $0.05777 |
| Opus 5 | $0.00031 | $0.02889 |
| Sonnet 5 | $0.00012 | $0.01155 |
| Haiku 4.5 | $0.00006 | $0.00578 |
Grade A, and why
wiki-ingest 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.
| `error` | 抓取失败 | 按 `reason` 字段判断回退路径（反爬走 web-access skill；gist/github raw 失败用 `curl`） | How it starts
The opening of the file, as written. The whole thing — 391 lines — stays where its author put it; the contents beside it link to each section on GitHub.
wiki-ingest
把外部内容（URL / 文件 / 粘贴文本）摄入当前 corpus：fetch → 写 wiki → 反链 → CLI 关账。
项目/domain skill 可以把 finished package、课程单元或客户材料路由到这里；它只补充来源分类、命名和验收规则，不重写 fetch/archive/wiki/lint/state 流程。
When to trigger
- 用户发来 URL / 文件路径 / 粘贴一段外部内容
- 用户说"ingest 这个"、"存一下"、"整理进知识库"、"收藏"、"归档一下"
- 用户发公众号链接（
mp.weixin.qq.com）且希望整理
不要触发：
- 对话中的洞察存回 →
wiki-fileback - 只是查询已有内容 →
wiki-query
分工边界（重要）
| CLI 负责（自动） | LLM 负责（你） |
|---|---|
| URL 查重（fetch 内建） | 主语识别 / Notability 判断 |
| 抽 publishDate / 写 frontmatter | Compiled Truth 写作 |
死链检测（ingest check） |
Timeline 措辞、反链选择 |
刷 corpus/index.md 受控区（sync） |
建反链时的 timeline 追加 |
写 corpus/log.md（record --log） |
一句话归纳本次 ingest 做了什么 |
| 推进 ingest state machine（record） | 决定哪些主语该建独立页 |
_INDEX.md / root index / doctor（sync） |
— |
| frontmatter 合规 / 死链 / 孤岛（lint） | — |
铁律：能让 CLI 做的就让 CLI 做。 不要手动 Edit index.md / log.md，不要先跑 lorekit search 做查重（fetch 已做）。
6 步流程
1. Fetch — lorekit fetch <url>
对任何 URL 统一调 lorekit fetch，按站点路由抓正文 + 图片，落 _工作台/收件/fetch/<slug>/。stdout 是单行 JSON：
| status | 含义 | 下一步 |
|---|---|---|
ok |
抓取成功 | 读 markdown 字段、按需读关键图。fetcher 自动写 status:started, stepsDone:['fetch'] |
duplicate |
URL 之前已 ingest 过 | 读 duplicate.path 看已有页，跟用户确认覆盖/追加/取消；要重抓加 --force |
in_progress |
上次 ingest 中断 | 读 ingestState.stepsDone + nextStep，从下一步继续，不要重抓 |
error |
抓取失败 | 按 reason 字段判断回退路径（反爬走 web-access skill；gist/github raw 失败用 curl） |
unsupported |
站点不直接处理（lark / x / pdf） | 按 suggest 字段用对应工具 |
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 · 391 lines · 62 tokens per session scan A 59af026e26a7
wiki-ingest is a skill published in the GitHub repository GYF0311/lorekit (5 stars, last pushed 1mo ago), licensed MIT. It adds 62 tokens to every session and 5,777 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
wiki-ingest
Ingest exactly one research paper or source document into an open-llm-wiki vault. Use when the user explicitly asks to add, ingest, process, or publish a paper into the wiki. The workflow parses the source, drafts a source page, runs independent QA, publishes only after the quality gate passes, extracts and normalizes…
openkb
Use when the user asks about content in their OpenKB knowledge base — research topics, concepts compiled from their documents, cross-document synthesis — or mentions openkb, an .openkb/ directory, or a wiki/ tree generated by openkb. The user may invoke you from any working directory; the active KB resolves via openkb…
mindbase
Maintain a persistent AI-tended research wiki. Activates when the user wants to ingest a source (PDF, URL, article, paste), capture a thought, build the wiki context, query what's already known, lint for orphans/contradictions, or maintain page conventions. Phrases include "add this to my wiki", "today I decided…"…
ingest-l1
L1 analysis loop for the abapwiki knowledge base: for each batch it launches the abap-analyzer sub-agent in parallel, then the adversarial judge abap-deepcheck (separate session), applies only the analyses that pass the fail-closed gate, and commits. Resumes exactly after an interruption. Use this skill to document…
wiki-ingest
Digest a course source file into the wiki. This skill should be used when the user says "ingest", "消化", "process this PDF/slide", or drops a course file into raw/ and wants it turned into wiki pages. Reads the source, deduplicates via manifest, creates source + concept pages, and finds cross-course connections.
wiki-diagram
Add Mermaid diagrams to wiki pages to aid understanding. This skill should be used during ingest/review when a concept involves a process, architecture, sequence, hierarchy, or comparison that is easier to grasp visually, or when the user explicitly asks to "画图"/"配图"/"diagram"/"visualize" a concept.