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 Songhonglei/better-agent-skills --skill multica-squad-load-balancergit clone --depth 1 https://github.com/Songhonglei/better-agent-skillsWrote 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/songhonglei/better-agent-skills/multica-squad-load-balancer)<a href="https://agentmods.dev/skills/songhonglei/better-agent-skills/multica-squad-load-balancer"><img src="https://agentmods.dev/badge/skills/songhonglei/better-agent-skills/multica-squad-load-balancer/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/songhonglei/better-agent-skills/multica-squad-load-balancer"><img src="https://agentmods.dev/badge/skills/songhonglei/better-agent-skills/multica-squad-load-balancer.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 | $0.00087 | $0.01475 |
| Opus 5.5 | $0.00035 | $0.00590 |
| Sonnet 5.5 | $0.00017 | $0.00295 |
| Haiku 4.5 | $0.00009 | $0.00147 |
Grade A, and why
multica-squad-load-balancer 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 6d 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Multica 小队负载均衡
- Version: 1.0.0
- License: MIT
- Author: Evan Song · github.com/Songhonglei
- Repository: https://github.com/Songhonglei/better-agent-skills
队长和成员安装同一份 Skill。使用已有 Multica CLI 登录及运行环境身份,仅依赖 python3 标准库;实际派单需要 Linux/macOS 等支持 fcntl 的 POSIX 环境。无额外数据库或业务系统依赖,不需要共享 Token。项目负责人应为 Agent,且在同一工作区领导唯一未归档小队;Issue 应指派给该小队。直接指派个人的 Issue 按原业务执行,不强行转换。
每次处理小队 Issue 时先读取完整 Issue、触发评论和自己的 task-run。通过 is_leader_task、项目负责人和实时小队名单判断角色;不得凭 Agent 名称判断。命令中的脚本路径相对于本 Skill 目录,应使用安装位置的真实绝对路径,不复制脚本到业务项目。
队长
-
首次派单运行:
python3 <skill目录>/scripts/squad_balance.py plan --issue-id <当前Issue UUID> --task-run-id <自身 MULTICA_TASK_ID> python3 <skill目录>/scripts/squad_balance.py dispatch --issue-id <当前Issue UUID> --task-run-id <自身 MULTICA_TASK_ID>plan只读,dispatch在真实队长运行环境中重算并派单。CLI 的--profile、--server-url、--workspace-id参数须放在子命令前;通常沿用运行环境默认配置,禁止固定为创建者的 profile 或服务器。 -
调度根据当前项目负责人解析小队,排除队长、非 Agent、归档/不可用/离线/达到并发上限的成员,按
queued/dispatched/prepared/running/claimed/starting的任务数升序选人;并列按 Agent UUID 升序。跨项目任务也计入该 Agent 的负载。working是正常可接单状态,是否满载由活跃任务数与并发上限判断。 -
业务需要特定技能或角色时,从原 Issue 已授权要求确定条件。用重复参数
--required-skill-id <工作区 Skill UUID>、--member-role <小队角色>,或 Issue metadata 中的multica_load_balance: {"requiredSkillIds": [...], "memberRole": "..."}。两条命令使用相同条件;成员异质时必须先限制能力再比较负载,不凭空替换业务 Skill。 -
dispatched:脚本已经在原 Issue 发出真实 mention,保存决策标记并记 squad activity;不要再手工发一次。保持小队 assignee,不创建子 Issue,队长不代做业务。 -
no_action:已有成员 run/派单/终态时不重派。读实际结果,按原业务流程处理。failed:说明离线、容量、身份或不确定结果的真实卡点;不自行换人重试。脚本写入失败后先读评论和 runs 对账,不能直接删除私有意图文件重跑。 -
再次唤醒先辨别诉求。致谢、成员已交付、等待人工或系统自动推进时不派单。用户明确要求续跑或重新执行,先核对上一执行、副作用及未完成处;本版不自动重开已派 Issue,应由项目原有受控续跑机制处理。
成员
-
只执行原 Issue 已授权任务,不递归调度。优先从运行环境取
MULTICA_TASK_ID、MULTICA_AGENT_ID,缺失时先读取当前上下文并唯一核验,不能挑选他人的 run。 -
开始业务前运行:
python3 <skill目录>/scripts/squad_balance.py verify-worker \ --issue-id <当前Issue UUID> --task-run-id <自身run UUID> --agent-id <自身Agent UUID> \ --binding-file <业务目录中的私有执行绑定.json>
What ships with it
6 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.
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.
- 6d ago First seen · 65 lines · 87 tokens per session scan A 2d3069d23c8d
multica-squad-load-balancer is a skill published in the GitHub repository Songhonglei/better-agent-skills (3 stars, last pushed 6d ago), licensed MIT. It adds 87 tokens to every session and 1,475 once invoked, about $0.0003 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 2026-10-20.
Other skills, from other repositories
test-result-feedback
Close the acceptance loop for a tested change — consume the apc test TierResult, map EACH acceptance criterion to a pass/fail outcome from structured fields (never chat text), attach an evidence ref, and write it back to the task. A SUT red is a finding, never softened to green.
spec-intake
Turn one atomic feature/gap request into a dispatchable task in a single pass — create the task, write SPEC.md (auto-folded into linkedAssetIds server-side), add acceptance criteria, record provenance. Two-part epics are refused, not silently split.
prismer-apple-reminders
Apple Reminders via remindctl: add, list, complete.
github-ops
GitHub CLI for issues, PRs, code search, CI logs, releases, and API queries. Requires gh CLI and auth.
ov-kanban
Work through tasks on a kanban board stored in OpenViking (viking://agent/kanban/ /, one markdown file per task). Use when the user says "ov kanban", "kanban", "看板", asks to put a task on the board, hand off the current work, continue a task another agent started, or run an unattended task loop. A board holds many…
task-forest
Maintains a repo-local task forest or task DAG for the current workspace. Use when the user asks to initialize or update a task forest, close a session, summarize evolving project work, align a request with a global goal, track progress/history/deviations/todos, save or apply a task proposal, or export the…