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 HsuanYuLee/polaris --skill checkpointgit clone --depth 1 https://github.com/HsuanYuLee/polarisWrote 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/hsuanyulee/polaris/checkpoint)<a href="https://agentmods.dev/skills/hsuanyulee/polaris/checkpoint"><img src="https://agentmods.dev/badge/skills/hsuanyulee/polaris/checkpoint/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/hsuanyulee/polaris/checkpoint"><img src="https://agentmods.dev/badge/skills/hsuanyulee/polaris/checkpoint.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 | $0.00121 | $0.00717 |
| Opus 5.5 | $0.00048 | $0.00287 |
| Sonnet 5.5 | $0.00024 | $0.00143 |
| Haiku 4.5 | $0.00012 | $0.00072 |
Grade A, and why
checkpoint 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 1mo 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.
What it actually says
Checkpoint
checkpoint 保存、恢復、列出長 session 的工作狀態，讓 context compression 或換 session
後能接回 branch、ticket、phase、next action。
Contract
Checkpoint 是 session continuity tool，不是 planning、triage、或 implementation skill。 它只寫 workspace memory / timeline 類的 local state，不替下游 skill 決定 scope，也不自動切 branch。
Mode Routing
| User says | Mode | Reference |
|---|---|---|
checkpoint, 存檔, save checkpoint, save state |
save | checkpoint-save-flow.md |
resume, 恢復, resume checkpoint, 接回去 |
resume | checkpoint-resume-list-flow.md |
list checkpoints, 列出存檔, show checkpoints |
list | checkpoint-resume-list-flow.md |
Ambiguous input defaults to save.
Reference Loading
| Situation | Load |
|---|---|
| Any run | polaris-project-dir.md, session-timeline.md |
| Save | checkpoint-save-flow.md, checkpoint-carry-forward-flow.md |
| Resume / list | checkpoint-resume-list-flow.md |
Hard Rules
- Save must capture branch, git status summary, ticket, todo disposition, phase, next action, and recent timeline context.
- 回報 checkpoint saved 前，carry-forward validation 必須通過。
- Resume verifies the checkpoint branch exists before suggesting a switch.
- 未取得使用者明確指示前，不切 branch、不 stash、不 rewrite memory。
- 不可默默丟掉 pending items；每項都要標記 done、carry-forward、或 dropped。
Completion
Save 回傳 branch、ticket、phase、next action、resume phrase。Resume 回傳 checkpoint timestamp、branch、ticket、可取得時的 current status、next action。List 回傳 recent checkpoint rows。
Step 2.5 — L2 Deterministic Check: cross-session-carry-forward
Save mode 的 carry-forward gate 由 checkpoint-carry-forward-flow.md 執行；必須呼叫
.claude/skills/checkpoint/scripts/check-carry-forward.sh，通過後才可回報 checkpoint saved。
What ships with it
7 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.
- references/checkpoint-carry-forward-flow.md 1.2 KB
- references/checkpoint-resume-list-flow.md 1.3 KB
- references/checkpoint-save-flow.md 1.7 KB
- references/polaris-project-dir.md 1.3 KB
- references/session-timeline.md 3.9 KB
- scripts/check-carry-forward.sh 14 KB runs code
- scripts/polaris-timeline.sh 7.4 KB runs code
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.
- 1mo ago First seen · 79 lines · 121 tokens per session scan A 33a368e18c47
checkpoint is a skill published in the GitHub repository HsuanYuLee/polaris (5 stars, last pushed 2d ago), licensed MIT. It adds 121 tokens to every session and 717 once invoked, about $0.0005 per session on Opus 5.5. 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
obsidian-workout-export
Export workout data to an Obsidian-formatted Markdown file. Always trigger immediately when the user types "owd". Also trigger on "export workout", "obsidian workout", "save workout to obsidian", or related requests to export workout data into Obsidian. Read the current conversation first before searching prior…
docs-driven-workflow
Use when starting or finishing any coding/editing task in a project that follows a documentation-first AI collaboration workflow — declare scope before editing, log every change afterward with a changelog entry, scaffold a brand-new project with this discipline (AGENTS.md/README/docs skeleton), or audit and reorganize…
agent-self-scheduling
Schedule AI agent runs with cron, loops, or external clocks while avoiding unsafe tight autonomous timers.
aident-loadout
Run multi-step work across apps the user connected in Aident Loadout: live action schemas, connection checks, cost preflight, and approval before side effects.
accint-commitments
Triage acc's open promises and close them with honest real-world verdicts via accact(runtime="outcome").
scenario-zbrush-expert
Use when an agent drives Maxon ZBrush 2026 for any task (sculpt in ZBrush, DynaMesh, ZRemesher, UVs, export, polypaint, posing, 3D print, clean up an AI mesh in ZBrush), when writing ZBrush Python (zbrush.commands) or ZScript, when the ZBrush bridge is not answering, a palette path or set() does nothing, a scripted…