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 GGbond-bo/MemOmics-Agent --skill delegation-orchestrationgit clone --depth 1 https://github.com/GGbond-bo/MemOmics-AgentWrote 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/ggbond-bo/memomics-agent/delegation-orchestration)<a href="https://agentmods.dev/skills/ggbond-bo/memomics-agent/delegation-orchestration"><img src="https://agentmods.dev/badge/skills/ggbond-bo/memomics-agent/delegation-orchestration/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/ggbond-bo/memomics-agent/delegation-orchestration"><img src="https://agentmods.dev/badge/skills/ggbond-bo/memomics-agent/delegation-orchestration.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 | $0.00609 | $0.50939 |
| Opus 5.5 | $0.00244 | $0.20376 |
| Sonnet 5.5 | $0.00122 | $0.10188 |
| Haiku 4.5 | $0.00061 | $0.05094 |
Grade A, and why
delegation-orchestration 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 — 1,334 lines — stays where its author put it; the contents beside it link to each section on GitHub.
子代理委派编排与回执处理
配套文件：
-
📎
scripts/extract_llm_json.py— 健壮 JSON 提取器（裸引号/尾随逗号/截断/容器递归全能）。合并子代理输出时直接跑它，别手写解析。 -
📎
scripts/verify_batch_replay.py— 回执重放判定器：把 §二 的 5 项检查一次跑完（容器形状预检 / 逐字段归一 diff / 独有串探针 / 闭环不变量 / 交付物指纹），返回码 0=ALREADY_MERGED、3=NEW_WORK。收到回执先跑它，再决定动不动手。 -
📎
scripts/audit_store_field_health.py— 判为重放后顺手跑的字段健康审计（逐字段长度分布 + 短字段清单 + 合法短值白名单），返回码 1=抓到真缺陷。见 §二 ⓪-d。 -
📎
templates/delegation-brief.md— 派发 brief 模板（复制填空即用）：目标 JSON 绝对路径 + 增量落盘契约（每写完一项立即回写）+ ≤5 篇/人 +fold -s -w 700读法 + 逐字引文构造法（t[i:j]切片，⛔ 不手打）+ 字段 schema 与最短字数 + 综述类固定填法 + 降级口径（先交部分 + 点名未完成）+ 完成判据。新派发前先复制它、⛔ 不要临场手搓 brief —— §一/§一-b/§四-b 记录的两次失败（6 篇零落盘、草稿只留内核变量）都出自即兴 brief。 -
📎
scripts/probe_replay_nonjson.py— 非 JSON 形态重放判定探针（一次跑完）：find 先行（⛔ 单层ls的 MISSING 不是证据）→ 落盘件 mtime 链 → 交付物覆盖计数 →find -newermt反向封闭 → 指纹（sha/字节权威 + 行数标wc -l口径）→ 台账「派发行 / 核销行」分离 → KB YAML 健康（先 print 键再扫）。返回码 0=ALREADY_REPLAY、3=NEW_WORK。见 §二 ⓪-j / ⓪-k。 -
📎
scripts/replay_reply_lint.py— 回放轮回复的「发前自查」linter（跑一下，别凭感觉）：扫 6 类正文违规（挂起/ 幻影路径复述 / 去重假承诺 / 序号泄漏（含标题行）/ 邀约句 / 疑似第五块），返回码 1=有违规（按提示整段删后重跑）、0=可发；2026-10-11 起再含长度门（正文 >1200 字符 或 >1 个 ```dsh-ui 围栏 ⇒ 同样返回 1，治「长正文被流切断 ⇒ 空回 ⇒ 同批白跑一轮」）；2026-10-13 起再含mermaid上限 0（规则冲突裁决见 §二 4-b-vi：SOUL 要求流程类配图，但回放轮四块骨架无图的槽位）+ 结构门（台账有[挂起]项时，正文出现编号 / 复选框 / 圈号清单即违规 —— 项名会被改写，字符串门追不上）。见 §二 4-b-iii / 4-b-iv / 4-b-vi。 🔴 2026-10-13 新增台账驱动的「项名门」（第七道门，治本）：加--ledger results/<sid>/notes.md， 项名从台账的[挂起]行读、不从记忆取 —— 见 §二 4-b-iii ⑦。命令：python scripts/replay_reply_lint.py --ledger results/<sid>/notes.md <草稿>。 -
📎
references/session-attribution-cross-check.md— ⓪-o 的第二条归属通道（不依赖任何缓存文件）：live log 整批不存在时，改用系统注入块定归属 —— 「会话产出资产」的相对路径逐条落回某个results/<sid>/＋ 「会话脚本目录 · scripts/」与候选 sid 的scripts/逐名相等；含ls -lt results/只触发调查不定归属、review/assets.json作为「我正在写哪个目录」的自身指纹，以及归属判完该如何选阶梯（本会话 → 重放；他会话 → 铁律 -5 隔离）。 -
📎
references/pending-item-leak-channel.md— 待决项泄漏的机制级根因（ⓠ 入口门之外的另一半）：task_plan.md的 Goal / Current Phase 区每轮唤醒被原样回灌 ⇒ 写进那里的待决项与#N序号会被逐词复述进回复（注入块 ↔ 回复逐词对照证据、写侧三条纪律、落笔前的机械自检）。见 §二 4-b-vii-b。 -
📎
scripts/audit_taskplan_injection.py— 注入区卫生审计（输入侧的对称门）：扫task_plan.md的 Goal / Current Phase / Phases 区，抓「待决项词汇」与「序号计数」两类会被每轮回灌、进而复述进回复的内容，返回码 1 ⇒ 就地清理那几行。replay_reply_lint.py管「写」（输出侧），本脚本管「读进来的是什么」（输入侧）—— 两者是同一件事的两半。见 §二 4-b-vii-b。
What ships with it
8 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.
- scripts/audit_store_field_health.py 6.0 KB runs code
- scripts/extract_llm_json.py 9.1 KB runs code
- scripts/probe_replay_nonjson.py 8.4 KB runs code
- scripts/replay_reply_lint.py 46 KB runs code
- scripts/verify_batch_replay.py 9.8 KB runs code
- skill.json 1.5 KB
- templates/delegation-brief.md 6.6 KB
- templates/replay-reply.md 1.5 KB
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 Changed · +164 lines · +67 tokens per session 92283a65a401
- 17d ago Changed · +56 lines 259e0e72120c
- 18d ago First seen · 1,114 lines · 542 tokens per session scan A c9adac7e794e
delegation-orchestration is a skill published in the GitHub repository GGbond-bo/MemOmics-Agent (30 stars, last pushed 5d ago), licensed MIT. It adds 609 tokens to every session and 50,939 once invoked, about $0.0024 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 2026-10-14.
Other skills, from other repositories
orbit-notion
Open Orbit briefing skill — selected by the Orbit pipeline when Notion is the user's only connected connector, or when the user explicitly scopes their daily digest to Notion. Pulls the past 24 hours of document edits, comments, mentions, and database row changes from the user's authenticated Notion connection and…
agentmail
Use your assigned AgentMail inbox to read email tasks, explicitly send or reply, and check delivery. Provided automatically by your inbox assignment.
Cortex
LifeOS memory, prior-session recall, and learning capture. USE WHEN capture this learning, document this learning, what did we learn recently, recent learnings, cortex, knowledge, knowledge base, search knowledge, what do we know about, archive, harvest, knowledge status, develop note, add to knowledge, add a people…
rowboat-spaces
How to work as a member of Rowboat Spaces, the team chat where people and agents work together. Load it once per session, before acting on any request that reaches you from Rowboat; it covers your setup, mentions and hand-offs, the tools, and files.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.