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.
git clone --depth 1 https://github.com/jzl-maker/EA-SKILLWrote 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/commands/jzl-maker/ea-skill/rec)<a href="https://agentmods.dev/commands/jzl-maker/ea-skill/rec"><img src="https://agentmods.dev/badge/commands/jzl-maker/ea-skill/rec/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/commands/jzl-maker/ea-skill/rec"><img src="https://agentmods.dev/badge/commands/jzl-maker/ea-skill/rec.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 | $0.00000 | $0.00661 |
| Opus 5.5 | $0.00000 | $0.00264 |
| Sonnet 5.5 | $0.00000 | $0.00132 |
| Haiku 4.5 | $0.00000 | $0.00066 |
Grade A, and why
rec 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
命令: /ea rec (恢复项目)
功能
最小代价加载项目当前状态。只读 state.md（≤50 行）+ context.md 摘要，按需查询详情。
触发
/ea rec [项目名称|路径]
加载策略（瘦身设计）
| 加载层 | 文件 | 何时读 |
|---|---|---|
| L0 默认 | <STATE_DIR>/state.md |
总是（≤50 行） |
| L1 类型 | <STATE_DIR>/project.json |
总是（< 20 行） |
| L2 上下文 | <STATE_DIR>/context.md |
总是，但只读摘要区段（见下） |
| L3 详情 | project-spec.md / problem-log.md / sessions/<id>.md / context.md 其余区段 |
按需（用户问到才读） |
⚠️ L2 不要整读 context.md
context.md 会随工程增长到几十 KB（实测某工程 77,832 B ≈ 20K tokens）—— 其中
「关键文件基线」的 sha256 列表和「增量学习记录」随每次 /ea record 变长。
整读等于把一次"轻量恢复"变成灌 20K tokens，与 ≤50 行 / 轻量恢复 的设计直接矛盾。
正确读法：只读 <!-- summary:begin --> 与 <!-- summary:end --> 之间
（芯片 / 工程与工具链 / 保护区清单）：
Grep: pattern="<!-- summary:(begin|end) -->" path="<STATE_DIR>/context.md" -n
Read: offset=<begin 行> limit=<end 行 - begin 行 + 1>
老工程没有标记时：按标题定位，只读 ## 芯片 到 ## 保护区清单 结束这一段，
不要因为找不到标记就退回整文件读取。其余区段等用户问到再按需读。
摘要输出格式
📂 项目恢复完成 — <项目名> (embedded)
芯片: GD32F407VET6 | 焦点: 显示模块
当前步骤: S3 — 验证中
下一步: <state.md 第 1 条 next-action>
详情命令: /ea stat -v
设计原则
- ❌ 不再一次性灌入 memory-log/project-spec/problem-log 全文
- ✅ rec 只回答「我现在该做什么」+ 工程上下文要点，详情按需查
相关文件
templates/state.md— state 模板templates/context.md— 工程上下文模板commands/stat.md— 详细状态查询（-v全景）
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 Changed · +10 lines c985966a0bc6
- 1mo ago First seen · 45 lines · 0 tokens per session scan A 7a5ded5d4acd
rec is a command published in the GitHub repository jzl-maker/EA-SKILL (9 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 661 tokens. 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 commands, from other repositories
runs
List live and persisted CCAM-launched Claude Code and Codex runs.
update-counts
Explicitly refresh public toolkit and committed marketplace source snapshots and numeric documentation.
orchestrate
You are the Project Orchestrator, the ultimate enforcer of the NASAB framework principles in all project workflows. You coordinate multi-agent work while ruthlessly enforcing quality gates, capability validation, and the principle of Patience (Pillar 6): perfect before advancing.
standup
Show a daily standup summary with completed, in-progress, and blocked tasks across all active epics.
iniciar
Inicializar Don Cheli en un proyecto nuevo o existente.
archivar
Archivar specs completadas y sincronizar delta specs con specs principales.