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/rongchangli/pocketskill/codex-save-tokennpx skills add RongchangLi/pocketskill --skill codex-save-tokengit clone --depth 1 https://github.com/RongchangLi/pocketskillWhat 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 | $0.00032 | $0.00551 |
| Opus 5 | $0.00016 | $0.00275 |
| Sonnet 5 | $0.00006 | $0.00110 |
| Haiku 4.5 | $0.00003 | $0.00055 |
Grade A, and why
codex-save-token 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 2d 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 — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Codex Save Token
Use this skill when the user wants Codex to reduce token usage without lowering engineering quality.
Core Principle
Spend tokens only on information that changes the next decision. Be concise, but do not skip context that is necessary for correctness, safety, or verification.
Working Rules
- Clarify only when a missing choice would materially change the outcome. Otherwise make a reasonable assumption and state it briefly.
- Search before reading. Prefer
rg,rg --files, targetedfind, and narrow path filters over opening broad directories or whole files. - Read slices, not dumps. Use
sed -n,nl -ba,head,tail,jq, or focused log ranges to inspect only relevant sections. - Avoid ingesting generated or bulky files unless they are directly relevant: lockfiles, minified bundles, build artifacts, long traces, notebooks, and large data files.
- Keep progress updates short and only when useful. Do not repeat the same status or restate the full problem.
- Prefer small, direct edits that match existing patterns. Avoid broad refactors, speculative abstractions, and unrelated cleanup.
- Verify with the narrowest meaningful command first. Escalate to broader test suites only when the touched behavior or risk justifies it.
- In the final answer, report the outcome, changed files, and verification in a compact form. Omit detailed command logs unless the user asked for them.
Tool Discipline
- Batch independent file reads with parallel tool calls when available.
- Set output limits on commands that may be verbose.
- Stop or narrow commands that start returning irrelevant output.
- For web or documentation lookup, browse only when current or source-specific information is required, then rely on primary sources and cite only what matters.
Conversation Discipline
- Preserve useful state through short summaries rather than re-reading the same files.
- After compaction or a long pause, resume from the latest concrete goal instead of replaying the whole history.
- When the user asks for a plan, keep it decision-oriented and short.
- When the user asks for code, implement rather than producing long speculative explanations.
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.
- 2d ago First seen · 45 lines · 32 tokens per session scan A 22d312d8b5fc
codex-save-token is a skill published in the GitHub repository RongchangLi/pocketskill (2 stars, last pushed 1mo ago), licensed MIT. It adds 32 tokens to every session and 551 once invoked, about $0.0002 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
hr-onboarding
A new-hire onboarding plan as a single page — first week schedule, buddy + manager intro, learning track, equipment checklist, and "you're set when…" outcomes. Use when the brief mentions "onboarding", "new hire", "first week plan", or "入职".
lark-im
飞书即时通讯：收发消息和管理群聊。发送和回复消息、搜索聊天记录、管理群聊成员、上传下载图片和文件、管理表情回复、发送应用内/短信/电话加急、发送和处理交互卡片（Interactive Card）、监听卡片按钮回调（card.action.trigger）。当用户需要发消息、查看或搜索聊天记录、下载聊天中的文件、查看群成员、搜索群、创建群聊或话题群、管理标记数据、管理 Feed 置顶（添加/移除/查询置顶会话）、管理标签数据、处理卡片回调时使用。.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
reflect
Review recent work, find repeated workflow patterns, and suggest reusable skills, agents, commands, config changes, or playbooks. Use when the user asks to learn from past sessions, improve recurring workflows, or identify what should be turned into reusable agent instructions.
mochi-remind
Handle due reminders — notify the user with natural language and mark them done.
organize-threads
猫猫辅助整理未分类 thread，分析标题和元数据，建议合适的标签。 Use when: 用户说"帮我整理"、"分类 thread"、点击整理按钮。 Not for: 删除/编辑标签本身。 Output: 按 thread 的标签建议列表。.