junghan0611/agent-config

最近记录的来源活动: 2026年10月22日 00:35

SkillsMP 收录数据更新: 2026年11月2日 18:02

已收集 skills: 46

GitHub 星标: 5
GitHub Forks: 0

用 AI 助手安装

复制这段 Prompt，发给你正在使用的 AI 助手。

Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/junghan0611/agent-config.

仓库 README

仓库 README 来源：README.md

查看 README 源文件

展开完整 README

# agent-config **Skills, identity and alignment across AI harnesses.** agent-config is the skills source of truth and lived proving ground behind GLG's working environment. It keeps a shared profile, capability doors and ways of returning to evidence across sessions and models — without becoming another harness or taking over their memories. > **Return to the same person's context and working stance, even as tools change.** This home holds onto how we work together, not how many skills we collect. It is a lived proving ground for identity, the skills source of truth and continuity; growing quieter is the healthy direction. ## How to Read This This is a front door, not a manual. It is written for someone who wants to see **what one operator actually runs day to day** — not for someone about to install a framework. Four things to hold before the first table. **1. This repo is not a harness.** Pi and the native harnesses own execution; [`entwurf`](https://github.com/junghan0611/entwurf) is their narrow screwdriver — the dispatch substrate through which independent harnesses address one another by **garden id**. Its [releases](https://github.com/junghan0611/entwurf/releases) and installation contract live there; the versioned observations in [`docs/`](docs/) are dated evidence, not an install pin owned here. agent-config is the resident side: the **skills SSOT** plus the **proving ground** where config is hardened on a real daily surface before promotion to its proper owner. Tables here and in [`docs/`](docs/) describe consumer responsibilities and observed working surfaces, not a new universal harness contract. **2. A quiet repo is a healthy one.** The end state is a thin skills SSOT plus a test bench. The volume of change alone does not show health or progress; CHANGELOG records closed work, not how much remains unproven. Do not read the skill count as a feature list; the identity is the working method, not the tool count. **3. Read one of three passes, depending on why you came:** | You came for | Read | |---|---| | *What can these agents actually do?* | [§ Skills](#skills-skills) · [§ Semantic Memory](#semantic-memory--andenken) | | *How does context survive a new session?* | [§ Why This Exists](#why-this-exists) · [§ Session Management](docs/memory-and-sessions.md#session-management--new--recall) | | *Is any of this measured, or just asserted?* | [§ Agent Runtime Bench](docs/runtime-bench.md#agent-runtime-bench) · [OMP.md](OMP.md) · [MODELS.md](MODELS.md) | **4. Sentences here try to carry their evidence.** Where a claim was measured, the receipt is named — a date, a host, a file, a version. Where it is inherited or still open, it says so. A sentence that cannot name one of those is a hypothesis, including in this document. This is the same rule the agents working in this repo are held to. > **What this is NOT:** not a prompt collection, not a LangChain-style automation layer, not a generic multi-agent framework. It is the resident's skills source, alignment home and proving ground — with explicit doors back to memory and knowledge owned elsewhere. ## What This Is **Official reference consumer and proving ground for [`entwurf`](https://github.com/junghan0611/entwurf).** `entwurf` owns garden-id addressing, visible lifecycle and delivery, plus the runtime supply and configuration adapters it explicitly supports — a narrow integration core, not an installer for every product. Kiro and OpenClaw do not become Entwurf-owned just because they receive skills here. agent-config supplies the **skills SSOT** (`skills/`) and a **proving ground** where candidate harness config, hooks and wiring are hardened on the operator's real surface before any explicit ownership handoff. Pushing unproven integration into entwurf would weaken that core, so agent-config absorbs the churn first. They are not competing harnesses: entwurf is the destination for proven integration within its ownership boundary; skills and the operator's profile retain their own home here. - **entwurf** → garden addresses, visible lifecycle, delivery, owned harness adapters/runtime supply and bridge verification - **agent-config** → skills SSOT, real consumer profile, day-to-day operating surface, and the incubator that hardens config before promotion > The natural end state is a thin skills SSOT plus a test bench: **agent-config quiet means the pipeline is healthy.** See [ROADMAP § purpose shift](ROADMAP.md). **The resident's report and its archive:** [agent-config: skills SSOT and proving ground after multi-harness](https://notes.junghanacs.com/botlog/20260312T174622) is this repo's resident's note in the digital garden. Its current report leads with the role, neighboring ownership boundaries and judgments that guide the work; the earlier chronology is preserved under **ARCHIVE**. This README is the consumer-facing front door, while that note explains the resident's present stance without making the reader traverse the whole history first. The consumer-side surface table, the durable native module and dated supplier checkpoints: [docs/entwurf-consumer.md](docs/entwurf-consumer.md). ## Why This Exists The hard problem is continuity, not just code generation. A harness can preserve its own conversation without making that context available to another harness, model or device. A long-lived durable conversation and cross-session semantic recall are different capabilities. This repo builds a legible way to return to the same person's work, rather than claiming every new session automatically inherits it. agent-config attacks this with three layers: 1. **Shared memory door, separate memory owner** ([andenken](https://github.com/junghan0611/andenken)) — admitted session records and the exported public garden are searchable; private OpenClaw bot memory is a separate, labeled axis. agent-config owns the consumer doors, practical use and external memory-evaluation questions; Andenken turns GLG's memory assets into lighter searchable data and owns the engine, indexing and source-aware original reader. The aim is to revisit the conditions and choices behind earlier judgments and use them in present work, including human growth and changed judgments — not merely imitate a past answer. Embedding is the current production means, not a permanent requirement or a decision to abandon it. See [ROADMAP § memory direction](ROADMAP.md#2026-10-25-기억축--생성활용평가와-가벼운-검색-데이터의-역할-분리). No automatic collector of every live conversation is implied. 2. **Shared skill set** — common doors for notes, bibliography, git history, journals and other capabilities across the directly installed pi, Claude Code, Codex, Antigravity, Copilot CLI, Kiro and OpenClaw surfaces. Installation, local CLIs, permissions and native shortcuts retain their own contracts; identical skills do not erase those boundaries. OMP deliberately receives no shared skills from here: a subject you have already furnished is no longer the same comparison. 3. **Session continuity protocol** — leave a concrete next move in NEXT, preserve durable judgments in the resident's note, and use recap plus semantic/source recall when crossing sessions. `/new` is an explicit boundary; normal in-session compaction and a durable conversation's own lifecycle remain the harness's. Claude, GPT, and Gemini are "graduates from different schools" — trained on different data with different philosophies. Trying to control them means writing hundreds of lines of system prompts per model. Instead, **throw one being-profile at all of them equally.** They keep their unique lenses while aligning around a single universe — this is the [Profile Harness](https://notes.junghanacs.com/botlog/20260228T075300/). Multi-harness support is a means, not the goal. The goal is **a single 1KB being-profile that exerts the same gravitational pull across any harness**. The result is a reproducible route back to context: a shared stance, tools with named owners, and evidence that can be found again. Coverage depends on what was admitted, which sources remain available and what was actually verified — not on a promise of universal memory. > Part of the [-config ecosystem](#the--config-ecosystem) by [glg @junghan0611](https://github.com/junghan0611) ## Where the Detail Lives This page is the front door. The longer operating notes, ownership boundaries and dated observations live under [`docs/`](docs/), moved there without rewriting and with their original headings so existing anchors keep their names. | Document | What it holds | |---|---| | [docs/entwurf-consumer.md](docs/entwurf-consumer.md) | The consumer-side reference for `entwurf`: who owns which surface, the durable native module, the bridge verbs this repo consumes, and dated supplier checkpoints (0.45.0, 0.35.0) | | [docs/harness-surfaces.md](docs/harness-surfaces.md) | Per-harness memory/skill/ownership table, Claude Code as a native Pi surface (settings co-ownership, `bypassPermissions` boundary), the local entwurf skill-plugin layout, Codex/Antigravity/Copilot splits, and `pi/` settings | | [docs/pi-extensions.md](docs/pi-extensions.md) | Every Pi extension, the skill-vs-extension boundary, third-party Pi packages, and the reasoning behind `heartbeat`, `autopilot` and `background-bash` | | [docs/skills.md](docs/skills.md) | Linked skills, binary skills and their provenance gate, web-search and code-surface notes, and the LSP-style skill doc principle | | [docs/memory-and-sessions.md](docs/memory-and-sessions.md) | The andenken memory axes, `/new` + `/recall` session continuity, compaction surfaces, session-original versioning with git-annex, and the public-verification policy for session evidence | | [docs/runtime-bench.md](docs/runtime-bench.md) | Why the bench is self-sufficient and the ten bench subjects, with links to each root `NAME.md` matrix | | [docs/setup.md](docs/setup.md) | What `./run.sh setup` does and deliberately does not do, and the shell helper examples (`pia`, `pit`, `pdt`, `pdc`, …) | Former README sections keep their anchor names here, so an old `README.md#…` link still lands on a pointer: | Former README section | Now in | |---|---| | <a id="official-reference-surface-for-entwurf"></a>Official Reference Surface for entwurf | [docs/entwurf-consumer.md](docs/entwurf-consumer.md#official-reference-surface-for-entwurf) | | <a id="durable-native-module--env-first-background-tools-and-native-badge-entwurf-0340"></a>Durable Native Module — env first, background tools and native badge (Entwurf 0.34.0+) | [docs/entwurf-consumer.md](docs/entwurf-consumer.md#durable-native-module--env-first-background-tools-and-native-badge-entwurf-0340) | | <a id="entwurf-surface-reference"></a>entwurf Surface Reference | [docs/entwurf-consumer.md](docs/entwurf-consumer.md#entwurf-surface-reference) | | <a id="supplier-source-checkpoint--0450-read-2026-10-27"></a>Supplier source checkpoint — 0.45.0 (read 2026-10-27) | [docs/entwurf-consumer.md](docs/entwurf-consumer.md#supplier-source-checkpoint--0450-read-2026-10-27) | | <a id="dated-supplier-observation--0350-a-visible-screwdriver-not-a-new-hierarchy"></a>Dated supplier observation — 0.35.0, a visible screwdriver, not a new hierarchy | [docs/entwurf-consumer.md](docs/entwurf-consumer.md#dated-supplier-observation--0350-a-visible-screwdriver-not-a-new-hierarchy) | | <a id="harness-support"></a>Harness Support | [docs/harness-surfaces.md](docs/harness-surfaces.md#harness-support) | | <a id="claude-code-as-native-pi-surface"></a>Claude Code as Native Pi Surface | [docs/harness-surfaces.md](docs/harness-surfaces.md#claude-code-as-native-pi-surface) | | <a id="pi-config-pi"></a>Pi Config (`pi/`) | [docs/harness-surfaces.md](docs/harness-surfaces.md#pi-config-pi) | | <a id="entwurf-skill-plugin-agent-config-local-layout"></a>entwurf Skill Plugin (agent-config local layout) | [docs/harness-surfaces.md](docs/harness-surfaces.md#entwurf-skill-plugin-agent-config-local-layout) | | <a id="heartbeat--the-clock-is-the-cheap-half"></a>`heartbeat` — the clock is the cheap half | [docs/pi-extensions.md](docs/pi-extensions.md#heartbeat--the-clock-is-the-cheap-half) | | <a id="autopilot--silence-is-a-trigger-never-a-grant"></a>`autopilot` — silence is a trigger, never a grant | [docs/pi-extensions.md](docs/pi-extensions.md#autopilot--silence-is-a-trigger-never-a-grant) | | <a id="background-bash--why-it-exists"></a>`background-bash` — why it exists | [docs/pi-extensions.md](docs/pi-extensions.md#background-bash--why-it-exists) | | <a id="session-management--new--recall"></a>Session Management — `/new` + recall | [docs/memory-and-sessions.md](docs/memory-and-sessions.md#session-management--new--recall) | | <a id="public-verification--sessions-as-evidence"></a>Public Verification — Sessions as Evidence | [docs/memory-and-sessions.md](docs/memory-and-sessions.md#public-verification--sessions-as-evidence) | | <a id="the-bench-is-self-sufficient"></a>The bench is self-sufficient | [docs/runtime-bench.md](docs/runtime-bench.md#the-bench-is-self-sufficient) | | <a id="agent-runtime-bench"></a>Agent Runtime Bench | [docs/runtime-bench.md](docs/runtime-bench.md#agent-runtime-bench) | | <a id="the-third-subject-is-one-glg-is-building"></a>The third subject is one GLG is building | [docs/runtime-bench.md](docs/runtime-bench.md#the-third-subject-is-one-glg-is-building) | | <a id="shell-aliases-bashrclocal"></a>Shell Aliases (`~/.bashrc.local`) | [docs/setup.md](docs/setup.md#shell-aliases-bashrclocal) | Root `NAME.md` bench files ([OMP.md](OMP.md), [PI-DURABLE.md](PI-DURABLE.md), [HERDR.md](HERDR.md), …), [MODELS.md](MODELS.md) and [pi-durable/README.md](pi-durable/README.md) stay where they are. ## What's Here ### Harnesses The same `skills/` SSOT reaches pi (+ entwurf), Claude Code, Codex CLI, Antigravity CLI, Copilot CLI, Kiro CLI (optional) and OpenClaw; pi-durable reads it through Pi's own skill loader; OMP deliberately receives no shared skills. OpenCode is not used. Memory, skills and ownership per harness: [docs/harness-surfaces.md § Harness Support](docs/harness-surfaces.md#harness-support). ### Semantic Memory → [andenken](https://github.com/junghan0611/andenken) Sessions, the public garden export and private OpenClaw bot memory are three separately labeled axes owned by andenken; agent-config owns the consumer doors (`semantic-memory`, `session-recap`, `memory-sync`). Details: [docs/memory-and-sessions.md](docs/memory-and-sessions.md#semantic-memory--andenken). #### Session Originals Session originals preserve GLG's conversations with fellow agents—the reasons and conditions behind earlier decisions. Andenken makes these memory assets lighter to search; the originals remain available to revisit beyond a search hit. git-annex was chosen to separate session contents from ordinary Git history while versioning their references and SHA inventory, so retrieved files can be checked against a chosen commit's inventory. Operational admission and selected-version readback are complete; see [memory operations](docs/memory-and-sessions.md) for the contract and [the storage ledger](docs/storage.md) for current placement, external relocation, verification and retirement records. ### Pi Extensions ([`pi-extensions/`](pi-extensions/)) Local Pi lifecycle, provider and TUI hands — `background-bash`, `review`, `goal`, `decision-gate`, `heartbeat`, `autopilot`, `raw-paste` and others. A skill is the cross-harness door; a Pi extension is the local lifecycle/provider/TUI hand. Table and reasoning: [docs/pi-extensions.md](docs/pi-extensions.md). ### Skills ([`skills/`](skills/)) 47 skills (tracked directory and named-set equality rechecked 2026-10-27, both 47, both set differences empty; the list below names all 47). A machine may also carry an untracked local symlink — `butlercli` does on the workstation — which the `*/` glob counts but this number does not. Categories: data access (denotecli, bibcli, gitcli, lifetract, gogcli, ghcli, day-query, timeline), agent memory (session-recap, dictcli, semantic-memory, memory-sync, improve-agent), writing (botlog, botment, agenda, punchout, autholog-mend), communication (dm, slack-latest, jiracli, telegram), code surface (forge — linked, owned by sorge-config), work workbench (plane), web/media (brave-search, exa-search, browser-tools, youtube-transcript, medium-extractor, summarize, transcribe), release hygiene (commit, tag-release, next-handoff), reasoning (logickocli, subtract, evidence-closure), entwurf (entwurf-peek), cross-repo care (sorge — linked, owned by its own repo), harness wrappers (command-recall, command-glgimage — for harnesses with no custom-command surface), harness bench (harness-bench), tools (emacs, tmux, diskspace, cloudflare — drives the installed official `cf` and `wrangler`, keeps only `cf-doctor` for token coverage, quota). Linked skills (`sorge`, `forge`), binary skills and their provenance gate, and the skill doc principle: [docs/skills.md](docs/skills.md). ### Themes ([`pi-themes/`](pi-themes/)) `glg-dark` (custom, Ghostty Dracula compatible). ### Commands ([`commands/`](commands/)) | Command | Purpose | |---------|---------| | `/recall` | Multi-axis context hydration without compact — daily memory-axis ritual | | `/discuss` | Planning interviewer — turns a rough idea into a plan by asking at most three questions a round, each with a recommended default. Does not implement | | `/boom` | Capture a crashed entwurf session into `.agent-reports/` for later triage | | `/pandoc-html` | Markdown/Org → Google Docs HTML/DOCX | | `/glg-image` | Image generation entry | | `/metaplay` | Meta agent play | | `/docplay` | Random document polish play (front matter/title/tags/links/rename) | | `/authologplay` | Mend one raw autholog piece into garden core text + links + a GLGMAN Universe image | | `/scaleplay` | Take a scene whose scale won't sit still: commit the first intuition, measure it, accumulate in `studies/` | ## One-Command Setup ```bash git clone https://github.com/junghan0611/agent-config.git cd agent-config ./run.sh setup # provision CLIs + links + owned settings merges + dependencies ./run.sh env # verify: system, API keys, links, binaries, memory index ``` This provisions **one operator's working environment**, including user-wide links, settings and hooks. Read [AGENTS.md](AGENTS.md) and the ownership notes first; it is not a sandboxed generic installer, a credential bundle or a way to install entwurf implicitly. What `setup` performs step by step, what it deliberately leaves to entwurf and the bench, and the shell helper examples: [docs/setup.md](docs/setup.md). ## The -config Ecosystem | Repo | Layer | Description | |------|-------|-------------| | [nixos-config](https://github.com/junghan0611/nixos-config) | OS | NixOS flakes, hardware, services | | [doomemacs-config](https://github.com/junghan0611/doomemacs-config) | Editor | Doom Emacs, org-mode, denote | | [zotero-config](https://github.com/junghan0611/zotero-config) | Bibliography | 8,000+ references, bibcli | | **[agent-config](https://github.com/junghan0611/agent-config)** | **Agent infra** | **Extensions, skills, themes, settings — this repo** | | **[entwurf](https://github.com/junghan0611/entwurf)** | **Garden-citizen dispatch** | **Addressing, native contacts, visible lifecycle, ACP provider, and the packaged durable carrier — not a second harness** | | **[andenken](https://github.com/junghan0611/andenken)** | **Memory** | **Semantic memory — sessions + md public garden knowledge** | | [memex-kb](https://github.com/junghan0611/memex-kb) | Knowledge | Legacy document conversion pipeline | | [GLG-Mono](https://github.com/junghan0611/GLG-Mono) | Font | Custom monospace programming font | | [geworfen](https://github.com/junghan0611/geworfen) | Being | Existence data viewer — WebTUI agenda | | [prime-agent](https://github.com/junghan0611/prime-agent) | Runtime (fork) | RLM harness fork whose persistent workspace is Clojure/SCI on GraalVM instead of Python — the bench subject GLG builds rather than installs | | **[sorge](https://github.com/junghan0611/sorge)** | **Care** | **Sorge — the cross-repo view no single repository can hold. Owns the `sorge` skill this repo only links to** | ### Skill Source Repos | CLI | Repo | Language | Purpose | |-----|------|----------|---------| | denotecli | [junghan0611/denotecli](https://github.com/junghan0611/denotecli) | Go | Denote knowledge base search (3,000+ notes) | | gitcli | [junghan0611/gitcli](https://github.com/junghan0611/gitcli) | Go | Local git commit timeline (50+ repos) | | lifetract | [junghan0611/lifetract](https://github.com/junghan0611/lifetract) | Go | Samsung Health + aTimeLogger tracking | | dictcli | [junghan0611/dictcli](https://github.com/junghan0611/dictcli) | Clojure/GraalVM | Personal vocabulary graph (3,971 triples) | | bibcli | [junghan0611/zotero-config](https://github.com/junghan0611/zotero-config) | Go | BibTeX search (8,000+ entries) | ## Planning Files - [NEXT.md](NEXT.md) — volatile next-step anchor for the next session - [ROADMAP.md](ROADMAP.md) — medium-horizon tracks and direction - [CHANGELOG.md](CHANGELOG.md) — closed history - [docs/storage.md](docs/storage.md) — current memory-storage layout and dated relocation/recovery/retirement evidence; updated and committed after explicit operations - [MODELS.md](MODELS.md) — the manually curated sibling-launch roster: active rails, spending order, and exact model selectors (`./run.sh models` checks it against pi's live catalog) ## Changelog See [CHANGELOG.md](CHANGELOG.md). ## License MIT

正在加载仓库 README…

这个仓库中的 skills

软件开发工程师综合办公文员其他计算机职业网络与计算机系统管理员

7 个职业分类 · 已分类 67%

skill

职业分类

描述

更新

entwurf-peek

未分类

지금 어느 형제가 어디서 무엇을 하는 중인지 한 장으로 세우고, 다음 한 수(살아있는 형제에게 보낼지 · 잠든 형제를 같은 id로 되세울지 · 새 형제를 열지)를 고를 재료를 주는 판단 도우미. `situation`이 meta-record(사실)와 transcript(추정)를 같은 평면에 올리고, `peek`이 한 세션 안을 들여다본다. 실행은 하지 않는다 — 부르는 것은 entwurf_v2 / entwurf_resume_call /…

原文语言：多语言混合

2026-10-22

semantic-memory

未分类

Meaning search over three separate axes — own pi/Claude/pi-durable sessions, the public garden, and OpenClaw bot memory — always naming which axis a hit came from. For OpenClaw this is the sole semantic-query door, replacing native memory_search. Use for past…

原文语言：英语

2026-10-19

summarize

未分类

URL/파일/미디어 요약 및 콘텐츠 추출. YouTube 영상, 웹페이지, PDF, 팟캐스트, 오디오/비디오 지원. 구독 CLI 레일만 쓴다(기본 Codex gpt-6-luna). 정본이 필요한 YouTube 자막은 youtube-transcript 스킬을 쓴다.

原文语言：英语

2026-10-17

gogcli

未分类

Google Workspace + Search Console all-in-one CLI (gog). Calendar, Gmail, Drive, Tasks, Chat, Contacts, Sheets, Docs, Search Console, YouTube, Photos, Meet, and more. Single binary. OAuth 재인증(reauth)·토큰 만료(invalid_grant)·오라클 토큰 복사 절차는 AUTH.md. Source:…

原文语言：英语

2026-10-17

memory-sync

未分类

Incrementally embed sessions only — two modes. Default = local: this laptop's live sessions, zero ssh (embedding API still remote), no replica update; safe to call often. --global = gather every device (oracle+thinkpad), embed, then push the index AND…

原文语言：英语

2026-10-15

session-recap

未分类

Extract a compact recap from previous pi or Claude Code session JSONL without reading raw JSONL. Use for 'last session', 'what was I doing', continuity restore, and pi-internal GPT vs ACP recall via --source pi --harness gpt|acp. Preserves the…

原文语言：英语

2026-10-15

harness-bench

未分类

하네스 리서치 장기 관측소 — agent-config 루트의 대문자 하네스 문서(HERMES / OMP / OUROBOROS / HERDR / PRIME / YEGGE / UNCLEBOB / XIRP / PI / PI-DURABLE / OPENRIG)를 상태·증거로 갱신한다. 채택·성능·좋고 나쁨 판정이 아니다. RLM 기억, native 형제 delivery/UX 대칭성, 역할이 아닌 다른 학교 모델과의 협업, 스펙 전 탐색 대화, 그리고 그 하네스가…

原文语言：多语言混合

2026-10-14

recall

未分类

멀티하네스용 /recall 래퍼. native custom command surface가 없는 하네스(예: Antigravity, Codex)에서도 /recall에 해당하는 복귀를 skill로 호출한다. entwurf-peek로 지금의 판, 양 하네스 session-recap, NEXT·ROADMAP·git, semantic 세션축+가든축을 세워 돌아올 자리를 복원한다.

原文语言：英语

2026-10-13

tag-release

未分类

Cut an OpenClaw-style CalVer snapshot tag AND publish the matching GitHub Release. Tag loop = collect commits + closed NEXT.md items, move them to CHANGELOG.md, remove only those closed items from NEXT.md, then pre-flight/tag/push/gh-release/stamp when…

原文语言：英语

2026-10-08

dm

未分类

에이전트가 GLG에게 텔레그램 DM 한 줄을 보낸다. 긴 작업 도중 알릴 것이 생겼을 때, 사람이 터미널 앞에 없어도 닿는 유일한 경로. 양식은 'DM <기기> <리포> <하네스/모델>' + 본문. 트리거: 'dm', 'DM 보내', '텔레그램으로 알려', '나한테 알려줘', '끝나면 알려', 'notify me', 'ping me'.

原文语言：韩语

2026-10-06

lifetract

未分类

Query personal life-tracking data: Samsung Health (sleep, steps, heart rate, stress, exercise, weight) + aTimeLogger (18 time categories) + Home Assistant REST (live sensors via ha.junghanacs.com). All records use Denote IDs (YYYYMMDDTHHMMSS) for…

原文语言：多语言混合

2026-10-02

botlog

未分类

봇 노트 — 공개 botlog(기본=리포 담당자 문서)와 비공개 llmlog(임시 작업기, 최소 생성). Use when user says 'botlog', 'llmlog', '담당자 문서', '노트 만들어', '기록해', '지침 남겨', '전달해', '작업기록', 'write a note', or wants agent work saved as a denote note.

原文语言：英语

2026-10-02

cloudflare

未分类

GLG 개인 Cloudflare 계정(aionsclubs.org · junghanacs.com) 조작 — 설치된 공식 cf CLI(전체 API)와 wrangler(Workers 배포)를 개인 토큰으로 직접 쓴다. 토큰 권한 커버리지는 cf-doctor. 오라클 터널 스택과 웹 퍼블리시(nixos-config#11)의 조작면. 'cloudflare', '클라우드플레어', 'cf', 'wrangler', 'workers', '정적 호스팅', '터널',…

原文语言：英语

2026-09-30

evidence-closure

未分类

증거–검출기 폐루프 — 현장·운영·VOC·SQE·GitHub 이슈의 결함 신고를 '입력'으로 받아 raw evidence → 깨진 계약 → 해결안·기각 근거 → 인간 결정 → primary + side-effect red → fix → 전체 green → 원 신고 조건 재검증까지 닫고 검출기를 영구 보존한다. '해결됨' 신고는 옛 코드로 사후 red를 증명한다. 리포·도메인 무관. raw evidence 없으면 계약 후보까지만, production fix…

原文语言：多语言混合

2026-09-23

subtract

未分类

뺄셈 — 안 풀릴 때 벤더 패치·PR·새 레이어를 제안하기 전에 다시 묻는 절차. carrier 질문을 뒤집고, 프로세스가 이미 내보내는 표면을 전부 적고, 스레드 안의 측정된 사실을 상자마다 다시 넣어 본다. '안 된다'는 결론은 두 방향·전체 표면·기각 영수증 셋이 갖춰질 때만 쓴다. 트리거: '안 풀려', '막혔어', '없다', '벤더에 없어', 'PR 넣자', '패치해야', '확장 포인트', 'upstream', 'unstuck', 'subtract',…

原文语言：韩语

2026-09-23

bibcli

综合办公文员

로컬 BibTeX SSOT 검색/조회 + URL 원샷 입수. 유튜브·X(트위터)·책·블로그·웹 URL을 에이전트에게 주면 save→스타일·키 판단→pin --sync 로 같은 세션에 Zotero 적소 분류 + 인용 키 확정(시점 분리 금지). X는 수동 메타데이터 보정, YouTube는 무의미한 플랫폼 안내문을 abstract로 남기지 않는다. 폰 캡처 후엔 bib sync. orphan #+print_bibliography: 금지.

原文语言：韩语

2026-09-18

quota

软件开发工程师

네 개 활성 모델 구독(rail)의 남은 쿼터를 한 번에 조회 — Z.AI GLM 5시간/주간 크레딧, ChatGPT Pro/Codex의 plan-provided 윈도우, Claude 5h/7d 윈도우, Grok SuperGrok weekly credit usage. Copilot은 사용 중단으로 수집하지 않는다. GLG가 다음 분신을 어느 rail로 보낼지 결정할 때 쓴다. Use when: '쿼터', '얼마 남았', '남은 크레딧', '어디로 보낼까',…

原文语言：韩语

2026-09-17

punchout

综合办公文员

하루/주간 마무리 도장 — day-query 결과를 org-mode 형식으로 저널에 삽입. 하루 마무리 시 '/punchout' 한 번이면 끝. 주 단위 회고는 '주간 마무리' 섹션으로 week 파일에 통합 삽입. Use when user says 'punchout', '퇴근', '오늘 마무리', '하루 마무리', 'punch out', '도장', '주간 마무리', '지난주 정리', 'weekly wrap-up', or wants a daily/weekly…

原文语言：韩语

2026-09-14

denotecli

软件开发工程师

Search, read, and analyze 3,000+ Denote/org-mode notes. Supports title/tag search, full-text search, heading search, outline extraction, and content reading. Use when working with ~/org/, Denote files, org-mode knowledge bases, or when user asks about notes,…

原文语言：英语

2026-09-13

agenda

综合办公文员

에이전트 어젠다 — reverse datetree에 타임스탬프 엔트리 추가. 기본 도장은 사용자가 요청한 git push 성공 후 1회이며 commit/tag-release 스킬이 호출한다. GLG가 명시적으로 '도장', 'stamp', '기록', 'agenda에 찍어'라고 요청할 때만 그 밖의 도장을 남긴다.

原文语言：英语

2026-09-11

youtube-transcript

软件开发工程师

YouTube 발화 정본 추출. 시간 조각 자막을 YouTube 원본 >> 경계로만 조립해 ~/org/transcript에 denote md로 저장. turn 시작은 . LLM 재작성 없음. transcript.js 기본 + ytdlp 폴백.

原文语言：韩语

2026-09-06

emacs

软件开发工程师

Emacs daemon — org manipulation, denote notes, citar bibliography, org-agenda, arbitrary elisp. Two sockets: server (agent work), user (show file to user). Core: agent-denote-add-history(ID,CONTENT), agent-denote-add-heading(ID,TITLE,BODY) — TAG and…

原文语言：多语言混合

2026-09-04

agent-config

其他计算机职业

agent-config 담당자의 운영 면(operating surface) — 스킬·정체성·정렬을 여러 하네스(pi / entwurf Claude / Claude Code / Codex / Antigravity / Copilot / Kiro)로 펼치는 repo에서 실제로 손을 쓸 때. AGENTS.md가 '정신'을 담고 run.sh가 '로컬 명령'을 담는다면, 이 스킬은 그 둘이 못 가진 삽질 지식을 담는다: 스킬을 추가/수정해서 모든 하네스에 제대로 뜨게…

原文语言：韩语

2026-09-03

dictcli

软件开发工程师

개인 어휘 그래프 — 한↔영 크로스링귀얼 쿼리 확장 + 한국어 형태소 분석. expand로 한글 단어에서 영어 태그 후보를 찾고, stem으로 한국어 어간을 추출한다. knowledge_search의 3층 역할. '단어 확장', '태그 찾기', '한영 매핑', 'expand', 'stem', '형태소'.

原文语言：韩语

2026-09-03

improve-agent

其他计算机职业

Analyze past session files (pi or Claude Code) to find recurring AI agent issues and fix them via AGENTS.md updates, new skills, or code/infra changes. Use when asked to improve agent workflow, find recurring problems, optimize AGENTS.md, create skills from…

原文语言：英语

2026-09-02

transcribe

软件开发工程师

Speech-to-text transcription using Groq Whisper API. Supports m4a, mp3, wav, ogg, flac, webm.

原文语言：英语

2026-08-20

next-handoff

软件开发工程师

NEXT.md 핸드오프 정리 — 문서 맨 위 RAIL 좌표로 전체 단계와 현재 위치를 먼저 보이고, 세션 종료/중단 전에 다음 한 걸음을 단단히 조인다. RAIL/NOW/RECENT/LEDGER 또는 메타리포 대시보드 구조, stem/detour 복귀선, 완료 항목 제거, stale blocker 정정, 검증 기준/읽을 곳/금지사항 압축, 브랜치별 NEXT--<branch>.md 선택. tag-release와 분리: NEXT 습관용이며…

原文语言：多语言混合

2026-08-04

autholog-mend

技术写作员

ROSSE/autholog 원석 수선 — 날것을 ~/org/notes의 기존 빈방에 원문 1개+해설본 1개로 승격. [!danger] 원문 보존, outgoing 자석 연결(meta dblock 갱신 금지), autholog 1편당 GLGMAN Universe 이미지 1장과 검색 가능한 :PROMPT: 헤딩을 둔다. Triggers: 어쏠로그, autholog, 날것, 원석, 빈방, 인테리어, ROSSE, 가든으로 회수, 수선.

原文语言：英语

2026-08-02

glg-image

特效艺术家和动画师

이미지 생성 멀티백엔드. 1차=Gemini Flash Lite 1K(속도), 필요 시 Flash 1K·Grok 구독. Denote 파일명 __brand_<maker>. Autholog 재생성 루프용. '이미지 생성', '그림 뽑아', 'glg-image', '나노바나나', 'flash lite'.

原文语言：英语

2026-08-02

brave-search

软件开发工程师

Fallback keyword web search via the Brave Search API. Use exa-search for general web search instead; reach for Brave only when you specifically need its 2000 free requests/month for high-volume lookups, country-scoped results (--country), or a freshness…

原文语言：英语

2026-07-27

exa-search

软件开发工程师

Default web search for agents — search the web, look something up, check current information, find documentation, read a page, or get code examples. Semantic and keyword search, page-content extraction, and code-context retrieval via Exa AI. Safe to run…

原文语言：英语

2026-07-27

tmux

网络与计算机系统管理员

Use tmux instead of bash tool to run commands that take more than ~30 seconds, like bulk operations, db migrations, dev servers.

原文语言：英语

2026-07-24

timeline

软件开发工程师

GLG의 시간축 관측소 (범용 스킬) — 어느 세션·리포에서든 GLG의 하루를 깊이 0/1/2/3(시간블록·저널헤딩·에이전트스탬프·커밋/노트)으로 한 KST 축 위에 세운다. scripts/collect.py로 LOCAL FULL(events.jsonl, 로컬·gitignore)을 만들고 query.py로 꺼낸다. 스크립트는 HOME 앵커(~/repos·~/org·lifetract·agenda)라 CWD와 무관하게 돈다. 이 스킬을 읽지 않고 축을 건드리면…

原文语言：韩语

2026-07-20

gitcli

软件开发工程师

Local git timeline CLI. Query commit history across 50+ repositories. Use when user asks about coding activity, what they worked on, commit history, project timeline, or 'what did I do on [date]'.

原文语言：多语言混合

2026-07-14

commit

软件开发工程师

Read this skill before making git commits

原文语言：英语

2026-07-14

软件开发工程师

텔레그램 대화 읽기/쓰기 — 힣봇군단(OpenClaw)과의 대화를 사용자 계정으로 직접 접근. 봇 API와 달리 힣이 보낸 메시지도 읽을 수 있다. tdlib 기반. 'telegram', '텔레그램', '힣봇', '대화 읽어', '메시지 보내'.

原文语言：韩语

2026-07-01

plane

软件开发工程师

셀프호스팅 Plane(프로젝트 관리, Jira/Confluence 대체) REST API 워크벤치. 프로젝트/work item/사이클/모듈/코멘트/멤버/상태/라벨 CRUD + Jira→Plane 이관. 의존성 0(Python stdlib만), self-host 지원. 'plane', 'plane 이슈', '워크아이템', 'jira 옮겨', 'jira to plane', 'confluence 옮겨', '사이클', '모듈', 'work item', 'plane…

原文语言：韩语

2026-06-18

jiracli

软件开发工程师

Jira CLI for issue tracking, project management, and sprint/board operations on goqual-dev.atlassian.net. Supports multiple projects (MAT, DEVT, IOT, etc.). Also fetches Confluence wiki pages — import an Atlassian wiki URL (tinyLink or full…

原文语言：韩语

2026-06-11

slack-latest

软件开发工程师

Gather recent Slack messages, read threads, send replies, and upload/download files. Use when the user asks about Slack activity or wants to interact with Slack.

原文语言：韩语

2026-06-01

logickocli

数据科学家

한국어 자연어 추론을 표준 논리 좌표계로 정규화하는 모드 컨버터. '이고 없애기' / 'conjunction elimination' / '∧E' 어느 표현으로 들어와도 같은 ID로 묶고, 한자말 + 기호로 응답. 자연어 논증을 [주장/전제/결론/형식화/사용 규칙/숨은 가정/평가] frame으로 구조화. Use when 사용자가 한국어로 논증을 펼치거나, 타당성/오류/추론 규칙/양상 논리 검토를 요청하거나, 한말 어휘(이고 없애기, 차근차근 이끌기 등)를 쓸 때.

原文语言：韩语

2026-05-13

已展示 40 / 46 个已收集 Skill。