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 skills add Leobai03/tc --skill tc-stategit clone --depth 1 https://github.com/Leobai03/tcWrote 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/leobai03/tc/tc-state)<a href="https://agentmods.dev/skills/leobai03/tc/tc-state"><img src="https://agentmods.dev/badge/skills/leobai03/tc/tc-state/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/leobai03/tc/tc-state"><img src="https://agentmods.dev/badge/skills/leobai03/tc/tc-state.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00115 | $0.01660 |
| Opus 5 | $0.00057 | $0.00830 |
| Sonnet 5 | $0.00023 | $0.00332 |
| Haiku 4.5 | $0.00012 | $0.00166 |
Grade A, and why
tc-state 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 9d 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 — 154 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TC State|创业状态
一句话说明
这是 TC 的“游戏存档”。今天做到一半可以保存;下次回来不用从第一关重新讲。
只处理五种请求
- 保存:把当前已经确认的创业判断和下一步写进本机存档。
- 恢复:读取某个项目最近一次存档,并从下一步继续。
- 列表:列出本机有哪些 TC 项目和存档。
- 报告:把同一项目的多次存档合成阶段报告。
- 证据导出:用户逐次明确同意后,导出不含项目名、联系人和完整事实清单的本机候选文件,交给
tc-knowledge人工复核。
不负责重新诊断、写文案或代替 tc-action 设计行动。发现问题定义还没有确认时,先回到 /tc 或 tc-diagnosis。
默认保存位置
~/.tc/projects/<项目名>/
├── current.md
├── current.json
├── sessions/
│ ├── 20260721T153000123456Z-标题.md
│ └── 20260721T153000123456Z-标题.json
└── reports/
└── 20260721T160000123456Z-阶段报告.md
所有文件默认只写在用户本机。不要声称作者可以看到这些文件。
保存前的边界
- 只有用户明确说要保存、记住或建立进度时才写文件。
- 保存前用一句话告诉用户将写入本机
~/.tc/。 - 默认不保存姓名、手机号、邮箱、账号密码、访问令牌、私聊全文、客户名单、合同原文和未公开商业机密。
- 需要保留人物关系时,用“合伙人 A”“客户 B”等代号。
- 用户要求保存敏感信息时,先提醒这是明文 Markdown,再让用户决定是否继续。
- 不自动上传到 GitHub、飞书、邮箱或其他网络服务。
- “同意保存状态”不等于“同意导出知识候选”;证据导出必须单独取得当次授权。
一份合格存档包含什么
只保存已经影响判断的内容:
problem_definition:这次真正解决什么;confirmed_facts:已经发生、能核对的事实;decision:当前选择;tradeoff:这个选择必须放弃或承担什么;rejected_directions:这次明确不做什么;assumptions:仍待市场验证的假设;next_action:下一项最小动作;success_metric:什么结果算有效;evidence:已经带回的用户原话、报价、付款、交付或毛利证据;next_skill:下次最可能进入/tc、tc-copy或tc-action中哪一个。
没有 problem_definition、decision 或 next_action 时,不要伪造;告诉用户还缺哪一项。
使用脚本
脚本只依赖 Python 标准库:
python3 scripts/tc_state.py save --payload /tmp/tc-state.json
python3 scripts/tc_state.py list
python3 scripts/tc_state.py restore --project 项目代号
python3 scripts/tc_state.py report --project 项目代号
python3 scripts/tc_state.py export-evidence \
--project 项目代号 \
--output /tmp/tc-evidence-candidate.json \
--consent
在 Agent 环境中执行时,先把结构化内容写入临时 JSON,再把路径传给 --payload。不要把用户原文直接拼进 shell 命令,避免引号和命令替换造成错误。
最小 JSON:
{
"project": "创业主线",
"title": "确定九十天唯一主线",
"source_skill": "tc",
"status": "active",
"problem_definition": "在不影响现有现金流的前提下,选出未来九十天唯一主线。",
"confirmed_facts": ["已有公开内容渠道", "当前同时处理多个方向"],
"decision": "只验证一个已有人付费的服务产品。",
"tradeoff": "暂时不新开赛道。",
"rejected_directions": ["同时测试三个新项目"],
"assumptions": ["现有咨询里存在可重复需求"],
"next_action": "整理最近十次真实咨询,统计重复问题。",
"success_metric": "七天内向五位同类客户正式报价。",
"evidence": [],
"next_skill": "tc-action"
}
What ships with it
2 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.
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.
- 9d ago First seen · 154 lines · 115 tokens per session scan A cf8358601477
tc-state is a skill published in the GitHub repository Leobai03/tc (22 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 115 tokens to every session and 1,660 once invoked, about $0.0006 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
knowledge-base-management
A lifecycle system for managing an Obsidian knowledge base, which is a folder of linked notes. It organizes raw material, AI-maintained wiki pages, and generated views into separate layers.
make-skill
A guided workflow for turning a conversation's solution into a reusable skill for coding agents.
agent-memory-mcp
A hybrid memory system that provides persistent, searchable knowledge management for AI agents (Architecture, Patterns, Decisions).
21-day-self-interview
A 21-night guided self-reflection routine in which an agent asks three questions each night, remembers the answers, and reflects on them at key points.
memory-log
A read-only timeline of changes to an agent's stored memories, including additions, corrections, downgrades, migrations, and rollbacks. It rebuilds the timeline from event records and current memory metadata.
obsidian-doc-structure
A read-only tool for inspecting the structure and properties of Obsidian notes. Obsidian is a note-taking app that stores linked Markdown files, and properties are fields such as status or date.