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 mikonos/LLM-GTD --skill updategit clone --depth 1 https://github.com/mikonos/LLM-GTDWrote 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/mikonos/llm-gtd/update)<a href="https://agentmods.dev/skills/mikonos/llm-gtd/update"><img src="https://agentmods.dev/badge/skills/mikonos/llm-gtd/update.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 | $0.01503 |
| Opus 5 | $0.00026 | $0.00751 |
| Sonnet 5 | $0.00010 | $0.00301 |
| Haiku 4.5 | $0.00005 | $0.00150 |
Grade A, and why
gtd-update 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GTD · update（状态更新 / 回报现实）
视角：David Allen，AI-native 改良。Update 处理的不是「新东西进系统」，而是现实已经变化：某个下一步做完了、等待的人回了、项目事实变了、日程细节确定了、承诺取消了。可信系统必须跟现实同步，否则它很快就会重新变成大脑里的负担。
加载与边界
- 清单边界和动作权限：先读
references/list-definitions.md。 - 涉及 hard landscape / 日历改动：读
references/capability-map.md。 - 涉及项目成果或下一步不清：按需读
references/clarify-decision-tree.md和references/natural-planning-model.md。 - 自动化边界：明确完成、明确等待回应、明确文本纠错可自动更新；多匹配、会删除整个项目块、日历事件无法唯一识别时问一句。
何时跑
- 用户说「做完了 / 搞定了 / 已买 / 已发 / 已确认 / 已经约好 / 对方回了」。
- 用户汇报项目进展：「我和家庭成员聊完了，已确认安排边界」。
- 用户纠正既有状态：「不是周三晚上，是周三中午 13:30 起飞」。
- 用户取消或降级承诺：「这个先不做了 / 不用跟了 / 这个项目关掉」。
与其他命令的边界
- 新输入 / 新承诺 →
capture/SKILL.md。 - inbox 里的未理清项 →
clarify/SKILL.md。 - 机械结构卫生（孤儿、重复、stalled 扫描）→
organize/SKILL.md。 - 此刻选做什么 →
engage/SKILL.md。 - 方向/项目是否还值得 →
review/SKILL.md。
Update 只做一件事：把用户刚报告的现实变化同步到 memory/gtd/ 和真实 hard landscape。
工作流
-
读相关清单再改：
- 默认先扫描
next-actions.md、projects.md、waiting-for.md。 - 若涉及日期 / 时间 / 地点 / 航班 / 会议 → 先读外部 calendar provider；全部不可达时读
calendar.md兜底。 - 若涉及项目支持事实 → 按需读
reference.md。
- 默认先扫描
-
判定 update 类型：
类型 判断口径 动作 下一步已完成 明确对应一条 next-actions删除该 next-action 行；若它挂项目，检查项目是否需要新下一步或已闭环 项目有进展 用户给了会影响下一步的事实 更新项目支持事实；移除已完成下一步；推导一个新的具体 next-action Waiting For 有回应 明确对应一条 waiting-for删除 waiting-for 行；把回应理清为 next-action / reference / 项目闭环 日程细节更新 日期 / 时间 / 地点 / 出发窗口变化 更新外部 calendar provider 事件；若找不到明确事件且信息完整则新建；不确定时问一句 承诺取消 / 不做 用户明确取消 删除对应行动或项目块；若只是暂缓且仍想保留，移到 someday-maybe 文本纠错 用户更正既有事实或措辞 最小替换对应清单行或项目支持材料 -
匹配规则：
- 只有一个清晰匹配项 → 直接更新。
- 多个相似项、会删除整个项目块、或日历事件无法唯一识别 → 先问一句短问题，不猜。
- 删除清单行是 GTD 闭环，不是删除文件；本命令可直接删除明确完成的行动行。
- 不把「已完成」留成打勾历史；活清单只保留仍需关注的承诺。
-
推进项目：
- 已完成行动属于某个 project 时，读该 project 块。
- 若期望成果已达成且没有仍需推动的下一步 → 删除整个 project 块。
- 若项目仍未完成 → 从新事实中起草一个当前最合适的 next-action，并把 project 的下一步指针更新到新行动。
- 如果新下一步需要用户价值判断或信息不足 → 不硬造，列出缺口并问一句。
-
日历 update 契约：
- 外部 calendar provider 可达时，它是 hard landscape；更新成功前不得声称已写入。
- 能唯一匹配现有事件 → update event。
- 找不到现有事件但有标题 + 日期 + 开始时间 → create event。
- 缺关键字段或可能重复 → 问缺失字段，不写猜测日程。
- 全部外部 provider 不可达 / 失败 → 写
calendar.md兜底并说明「未写入外部日历」。
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 · 93 lines · 52 tokens per session scan A b0fafda12426
gtd-update is a skill published in the GitHub repository mikonos/LLM-GTD (10 stars, last pushed 2mo ago), licensed MIT. It adds 52 tokens to every session and 1,503 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
give-me-tips
Explains any senpi tip in depth, including Tip: lines in the TUI. Use when the user asks about a tip, what a tipped feature does, or which tips they can see.
getting-started
Interactive post-onboarding tour that adapts to whatever data exists (calendar, Granola, or none). Use right after onboarding, or when the user says 'show me around', 'how do I start'. Also use proactively when the vault is < 7 days old. Not for the initial setup itself; use setup.
career-coach
Personal career coach with 4 modes: weekly reports, monthly reflections, self-reviews, promotion assessments.
resume-builder
Build resume and LinkedIn profile through guided interview.
daily-plan
Build today's plan from calendar, tasks, priorities and commitments, with smart scheduling suggestions. Use when the user says 'plan my day', 'what's on today', 'help me focus', or starts the morning. Also use proactively at the first session of the day. Not for reviewing a finished day; use daily-review.
career-setup
Initialize career development system (job description, ladder, reviews, goals).