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/lilmgenius/paperthin/macrothinknpx skills add LilMGenius/paperthin --skill macrothinkgit clone --depth 1 https://github.com/LilMGenius/paperthinWhat 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 | $0.00050 | $0.00846 |
| Opus 5 | $0.00025 | $0.00423 |
| Sonnet 5 | $0.00010 | $0.00169 |
| Haiku 4.5 | $0.00005 | $0.00085 |
Grade A, and why
macrothink 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 3d 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 — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Step back from the session's chosen path and ask several fresh reads what the current direction might be missing.
Goal
macrothink checks the live decision or direction in hand, not the artifact that describes it. It is for moments when the session may have inherited a framing from examples, wording, prior turns, or a first plausible answer.
The product is a spread of independent reads. Divergence is the signal. Convergence is reassurance, never proof.
Workflow
This skill is not model-invoked automatically. When a user explicitly invokes it, the invoked run may create the bounded fresh sub-sessions below; they are read-only inputs to this pass, not separate authority to change the plan.
- Name the current direction: the decision the session is about to keep building on.
- Strip the prompt to the underlying problem: goal, constraints, and known facts. Remove the session's own examples, suggested answer, preferred naming, and framing-specific wording.
- Fan out independent fresh reads of that stripped restatement: 2 to 5, default 3. Same model is allowed because this pass does not claim cross-model verification.
- Collect each read without correcting it toward the session's current direction.
- Classify each read against the current direction:
divergent-incompatible: challenges a premise the direction depends on.divergent-compatible: adds or reframes something useful without discarding the direction.convergent: independently lands near the current direction.
- Cluster before reporting: when several reads diverge in the same direction, name the shared root they point at — the pattern behind the divergences is the finding, and the individual reads sit under it as evidence. Do not report five specifics that are one gap seen five times.
- Report
divergent-incompatiblefirst, thendivergent-compatible, thenconvergent. Label convergence as reassurance only. - Return control to the main session. This skill is read-only and advisory; it does not rewrite the plan or pick the final answer.
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.
- 3d ago First seen · 52 lines · 50 tokens per session scan A 4f7808b62054
macrothink is a skill published in the GitHub repository LilMGenius/paperthin (999 stars, last pushed 14d ago), licensed MIT. It adds 50 tokens to every session and 846 once invoked, about $0.0003 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
foblex-flow
Authoritative guide for working with Foblex Flow (@foblex/flow) in the skill-map UI. Use whenever editing Angular code in ui/ that touches graph rendering — templates with f-flow / f-canvas / f-connection / fNode / fConnector / fDraggable / fZoom / fMarker directives; TypeScript importing from @foblex/flow…
mm-web-bridge
Партнёр louise в claude.ai — обсуждает идеи, ставит их под сомнение, проверяет актуальность в интернете перед решениями на внешних API/библиотеках, и оформляет self-contained промпты для её Claude Code в PowerShell. Use whenever louise обсуждает идею или фичу, просит собрать промпт/задание для PowerShell-Клода…
mm-init-project
Инициализирует или обновляет проект для mm-системы — создаёт passport.md в корне, копию в Obsidian, dashboard.md, handoff.md (скелет), project-instructions.md для claude.ai. Use when user says "оформи проект", "сделай паспорт", "init project", "/mm-init", "/mm-init-project", "обнови паспорт", "регистрирую проект".…
mm-save-session
Закрывает текущую Claude Code сессию — сохраняет лог в Obsidian/Claude/Sessions/, обновляет project note, обновляет INDEX.md, перегенерирует handoff.md (для Project Knowledge claude.ai) через /mm-handoff. Если в проекте есть GSD (.planning/ или .gsd/) — также вызывает /gsd-pause-work для technical state в…
mm-resume
Восстановить контекст в начале сессии или после /clear. Читает passport.md + dashboard + последнюю сессию + git status + (если есть GSD) полный контекст планирования (.planning/ для v1 или .gsd/ для v2 — STATE, ROADMAP, текущий phase CONTEXT/PLAN, HANDOFF.json). Выдаёт компактную сводку «где мы». Use when user says…
mm-clear-gate
Гейт перед /clear — read-only проверка «всё ли записано», вердикт из exit code скрипта, а не из реплики модели. Проверяет свежесть GSD-handoff в репозитории проекта, свежесть vault-файлов для Project Knowledge, состояние vault-репо, дерево проекта, запушенность, лог сессии, плюс validate.health как предупреждение. Два…