Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add theseussss/roamfound/plugin install jiacong-flowWrote 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/theseussss/roamfound/smarter-project)<a href="https://agentmods.dev/skills/theseussss/roamfound/smarter-project"><img src="https://agentmods.dev/badge/skills/theseussss/roamfound/smarter-project.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 5.1 | $0.00052 | $0.09358 |
| Opus 5 | $0.00026 | $0.04679 |
| Sonnet 5 | $0.00010 | $0.01872 |
| Haiku 4.5 | $0.00005 | $0.00936 |
Grade A, and why
smarter-project 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 8d 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 — 411 lines — stays where its author put it; the contents beside it link to each section on GitHub.
smarter-project
⚠️ 执行规约（非知识文档）。读完要跑脚本、要写 scratch，不是只"理解"。
你在一个项目中和用户反复对话，讨论会产生大量认知碎片。如果这些碎片散落在对话历史里，下次开会话时它们就消失了。smarter-project 解决的就是这个问题——把讨论中产生的认知组织成可持久、可检索、可继续推进的结构。
做法是话题化：每一个值得追踪的讨论方向，都开一个话题目录（topics/NNN_简称/），里面放三件套——
| 文件 | 性质 | 承载 |
|---|---|---|
scratch.md 讨论卡 |
每轮追加，不删不清 | 原始讨论过程——"当时怎么想的" |
card.md 话题卡 |
唯一真理源 | 从讨论中提炼的结构化结论 |
tasks.md 任务卡 |
可选 | 从话题派生的执行步骤 |
基础话题卡（base_topic: true）可选生成第四组件 template.md，承载建构说明、可重复单元模板与地图渲染约定。普通话题默认三件套，需要时通过 --with-template 生成。详见 references/base-topic-templates.md。
话题文件夹里也允许放额外文件（诊断报告、方案草稿等），但必须在 scratch 对应条目中索引，确保能追溯来源。
每个文件顶部有 YAML frontmatter（topic_id、status、parent、root 等）。card.md 以 frontmatter toc: 为唯一结构源（格式与约束详见 §2.1 话题），正文 ## 📑 本卡目录 只是由脚本渲染的只读视图。status 标记生命周期阶段，完整流转判据见 references/topic-lifecycle.md。
全部状态落在 markdown 文件中，不做数据库，打开编辑器就能审计。Python 脚本（scripts/*.py）负责建话题、校验结构、生成可视化等操作——AI 必须调用脚本，禁止手工 mkdir + 手写文件替代，因为脚本才能保证格式一致。
现有 HTML 仪表盘（dashboard.html、role_manager.html、_tree.md 可视化）是未来统一 GUI 的理念基石。
下面按实际执行流来组织。<skill> = ${CLAUDE_PLUGIN_ROOT}/skills/smarter-project。
| § | 内容 | 管什么 |
|---|---|---|
| 1.1 | 总纲 | 全局设计原则 |
| 1.2 | 会话启动 | 判断是否建项目、体检、首段输出 |
| 1.3 | 每轮循环 | 焦点检查 → 角色 → 执行 → 写 scratch |
| 2.1 | 话题 | 三件套的建、改、迁移、收敛、额外文件 |
| 2.2 | 记录机制 | scratch / log / tasks.md / Trace（追踪）话题四层分工 |
| 3.1 | 角色 | 蒸馏高手，视角一以贯之 |
| 3.2 | 初始化 | 套餐 + 基础话题卡集 |
| 4 | 审查 | 结构对齐与一致性检查 |
§1.1 · 总纲
三条全局原则，贯穿后续所有章节。
md 唯一真理源。 讨论与结论状态在 markdown 文件的 frontmatter 和正文中；项目运行态、缓存和派生视图默认落在 .jiacong/ 下，可删除重建。不做数据库、不做双写。旧 .claude/ 只作 Claude hook 配置或 legacy fallback。
结构权归用户。 结构变更——改 TOC、增删章节、新建话题、偏离套餐——由 AI 提议、用户审批。内容填充（在已有章节内写正文）由 AI 动笔、用户校对。用户未明确批准的提议一律视为拒绝：有效批准词是"同意/通过/可以/就这样/对/行/OK/去做/开搞"；含糊回复（嗯/哦/知道了/好）须追问确认。
脚本优先。 凡有对应脚本的操作，AI 必须调用脚本。手工替代是最常见的执行偏差来源。
§1.2 · 会话启动
会话第一轮有一个额外判断：这个对话需不需要项目管理？
如果用户只是随手问个问题、做个一次性查询，skill 不介入，直接回答。如果任务涉及多轮协作、需要持久化认知（论文写作、方案设计、代码项目等），就需要建档。不确定时，问用户。
建档前必须先做入口分类：当前打开的文件夹究竟是标准 .git 项目根、workspace 容器、workspace 内 worktree、无 git 历史文件目录、多 repo 工作区，还是半建档旧结构。分类规则见 references/init-project.md §1.5。未完成入口分类前，不直接运行初始化脚本。
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- ALGORITHM.md 8.4 KB
- ARCHITECTURE.md 17 KB
- docs/settings-hook-example.json 565 B
- references/base-topic-templates.md 8.0 KB
- references/focus-switch.md 5.5 KB
- references/init-project.md 23 KB
- references/proposal-bridge.md 2.4 KB
- references/review-ops.md 6.1 KB
- references/roles-ops.md 7.5 KB
- references/topic-lifecycle.md 16 KB
- references/vision-frameworks.md 7.5 KB
- roles-library/_index.md 3.5 KB
- roles-library/academic/philosopher-of-operationalization.md 2.6 KB
- roles-library/frameworks/first-principles-thinking.md 18 KB
- roles-library/frameworks/toulmin-model.md 7.1 KB
- roles-library/methodology/concept-formation-analyst.md 2.9 KB
- roles-library/methodology/construct-validity-theorist.md 3.0 KB
- roles-library/methodology/domain-methodologist.md 2.6 KB
- roles-library/writing/methodological-writing-editor.md 2.9 KB
- scripts/_lib/__init__.py 96 B runs code
- scripts/_lib/autofix.py 402 B runs code
- scripts/_lib/common.py 1.6 KB runs code
- scripts/_lib/data.py 12 KB runs code
- scripts/_lib/entrypoints.py 3.3 KB runs code
- scripts/_lib/flow.py 1.3 KB runs code
- scripts/_lib/guard.py 24 KB runs code
- scripts/_lib/journal.py 5.8 KB runs code
- scripts/_lib/store.py 34 KB runs code
- scripts/_lib/structure.py 1.3 KB runs code
- scripts/_lib/topics_loader.py 1.9 KB runs code
- scripts/_lib/view.py 14 KB runs code
- scripts/active_topics.py 4.4 KB runs code
- scripts/card_write.py 16 KB runs code
- scripts/check_links.py 3.5 KB runs code
- scripts/check_structure.py 3.0 KB runs code
- scripts/check_units.py 5.8 KB runs code
- scripts/dashboard.py 34 KB runs code
- scripts/flow_hook.py 16 KB runs code
- scripts/focus_breadcrumb.py 3.3 KB runs code
- scripts/health_check.py 7.5 KB runs code
- scripts/init_project.py 34 KB runs code
- scripts/render_maps.py 5.8 KB runs code
- scripts/role_manager.py 4.5 KB runs code
- scripts/topic_new.py 7.2 KB runs code
- scripts/tree_gen.py 8.5 KB runs code
- scripts/watcher.py 13 KB runs code
- templates/console.html.tmpl 57 KB
- templates/framework/framework_file.md.tmpl 568 B
- templates/framework/Framework_README.md.tmpl 1.1 KB
- templates/framework/README.md.tmpl 830 B
- templates/project/dashboard.html.tmpl 81 KB
- templates/project/project_entry.md.tmpl 7.1 KB
- templates/role/framework.md.tmpl 1.5 KB
- templates/role/manager.html.tmpl 13 KB
- templates/role/profile.md.tmpl 2.0 KB
- templates/topic/_tree.md.tmpl 2.6 KB
- templates/topic/card.md.tmpl 1.5 KB
- templates/topic/scratch.md.tmpl 958 B
- templates/topic/tasks.md.tmpl 903 B
- templates/topic/template.md.tmpl 1.6 KB
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.
- 8d ago First seen · 411 lines · 52 tokens per session scan A a3fb570dbc3b
smarter-project is a skill published in the GitHub repository theseussss/roamfound (5 stars, last pushed 2mo ago), licensed MIT. It adds 52 tokens to every session and 9,358 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 2026-08-31.
Other skills, from other repositories
recipe-create-meet-space
Create a Google Meet meeting space and share the join link.
workthreads
SpecStory Workthreads - a weekly work-thread rollup across a team's repos from SpecStory coding histories (any agent - Claude Code, Codex, Cursor, Gemini, and more). It groups the window's sessions into threads of work per project and labels each new / open / recently closed, so a lead sees what shipped, what is still…
atmos-config
Atmos root configuration: atmos.yaml discovery, precedence, deep merging, basepath, imports, minimal bootstrap, and routing to narrower Atmos skills.
story-readiness
Validate that a story file is implementation-ready. Checks for embedded GDD requirements, ADR references, engine notes, clear acceptance criteria, and no open design questions. Produces READY / NEEDS WORK / BLOCKED verdict with specific gaps. Use when user says 'is this story ready', 'can I start on this story', 'is…
autotask-creator
Rules for automation CRUD from the group-chat commander. The commander does not call mutation tools and does not edit cloud/autotasks files directly. It emits one or more top-level ... containers in its final text; the bus parses and applies them after the turn.
monorepo-management
Master monorepo management with Turborepo, Nx, and pnpm workspaces to build efficient, scalable multi-package repositories with optimized builds and dependency management. Use when setting up monorepos, optimizing builds, or managing shared dependencies.