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/konglong87/superpm/pm-prioritynpx skills add konglong87/superPM --skill pm-prioritygit clone --depth 1 https://github.com/konglong87/superPMWrote 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/konglong87/superpm/pm-priority)<a href="https://agentmods.dev/skills/konglong87/superpm/pm-priority"><img src="https://agentmods.dev/badge/skills/konglong87/superpm/pm-priority.svg" alt="Measured on agentmods" 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 | $0.00059 | $0.05539 |
| Opus 5 | $0.00030 | $0.02769 |
| Sonnet 5 | $0.00012 | $0.01108 |
| Haiku 4.5 | $0.00006 | $0.00554 |
Grade A, and why
pm-priority scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 655 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Preamble (run first)
bash "$(dirname "${BASH_SOURCE[0]}")/../../check-update.sh" 2>/dev/null || true
# 创建需求调研目录
mkdir -p docs/01-需求调研
# 检查是否有确认需求清单
if [ ! -f "docs/01-需求调研/确认需求清单.md" ]; then
echo "⚠️ 未找到确认需求清单"
echo ""
echo "建议先执行 /pm-clarify 细化需求"
echo ""
echo "您可以选择:"
echo "A) 执行 /pm-clarify 先细化需求(推荐)"
echo "B) 手动输入需求列表(快速模式)"
echo "C) 从需求池导入(执行过 /pm-pool)"
fi
跨 Agent 交互规则
当流程要求与用户交互时:
- 如果当前环境支持 AskUserQuestion,使用 AskUserQuestion(最佳体验)。
- 如果当前环境不支持 AskUserQuestion,必须用普通聊天消息提出同样问题。
- 一次只问一个问题。
- 提问后必须停止当前回合,等待用户回答(STOP and WAIT)。
- 不得在用户回答前生成文档、写入 docs。
- 已有 docs 文件不能替代本轮用户回答。
执行流程
步骤 1: 读取前置数据(主 agent)
如果有确认需求清单:
使用 Read 工具读取 docs/01-需求调研/确认需求清单.md
提取需求列表。
如果有需求调研报告:
使用 Read 工具读取 docs/01-需求调研/需求调研报告.md
提取初步需求列表和背景信息。
如果有市场调研报告:
使用 Read 工具读取 docs/01-需求调研/市场调研报告.md(如存在)
提取市场数据、竞品信息,为后续评分提供数据支撑。
如果没有前置文档:
进入快速模式,使用 AskUserQuestion 收集需求列表。
步骤 2: 选择排序模型(主 agent - 用户交互)
使用 AskUserQuestion 询问:
🎯 选择需求优先级排序模型:
A) RICE评分 - 综合Reach、Impact、Confidence、Effort(推荐) B) KANO模型 - 基于用户满意度分类需求 C) MoSCoW法则 - Must/Should/Could/Won't分类 D) 自定义权重 - 自定义评分维度
用户选择后,记录到变量 PRIORITY_MODEL
步骤 3: Subagent 并行预分析(v2.0 核心优化)
优化说明:
- 每个需求派发一个 subagent,读取所有前置文档,基于数据和所选模型生成初始评分建议
- 多个需求并行分析,总耗时等于单个需求分析时间
- 预分析结果包含评分 + 数据依据,用户只需确认或微调
- 大幅减少逐维度询问的交互轮次
3.1 构建预分析任务
对每个需求,构建一个 subagent 任务:
[
{
"task_id": "prescore_req_1",
"requirement": "{需求1名称}",
"model": "{PRIORITY_MODEL}",
"docs": ["docs/01-需求调研/确认需求清单.md", "docs/01-需求调研/需求调研报告.md"]
},
{
"task_id": "prescore_req_2",
"requirement": "{需求2名称}",
"model": "{PRIORITY_MODEL}",
"docs": ["docs/01-需求调研/确认需求清单.md", "docs/01-需求调研/需求调研报告.md"]
}
]
如果需求数量 > 10,提示用户:
⚠️ 需求数量较多({N}个),预分析可能需要较长时间
您可以选择: A) 全部预分析 B) 仅对核心需求预分析(前10个) C) 跳过预分析,直接手动评分
3.2 并行派发 subagent
同时派发所有 subagent:
# 并行派发 N 个 subagent
Agent 1: 预分析-{需求1}
Agent 2: 预分析-{需求2}
...
Agent N: 预分析-{需求N}
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.
- 3d ago First seen · 655 lines · 59 tokens per session scan A d6c33ab7d8eb
pm-priority is a skill published in the GitHub repository konglong87/superPM (60 stars, last pushed 21d ago), licensed MIT. It adds 59 tokens to every session and 5,539 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
deepsafe-scan
Preflight security scanner for AI coding agents — scans deployment config, skills/MCP servers, memory/sessions, and AI agent config files (hooks injection) for secrets, PII, prompt injection, and dangerous patterns. Runs 4 model behavior probes (persuasion, sandbagging, deception, hallucination). Supports LLM-enhanced…
douyin-upload-mcp-skill
通过抖音创作者平台(creator.douyin.com)发布视频和图文内容。首选调用 MCP 工具完成操作,MCP 无法满足时可运行 Skill 脚本兜底。禁止自行启动外部浏览器访问抖音。.
team-tasks
Coordinate multi-agent development pipelines using shared JSON task files. Use when dispatching work across dev team agents (code-agent, test-agent, docs-agent, monitor-bot), tracking pipeline progress, or running sequential/parallel workflows. Covers project init, task assignment, status tracking, agent dispatch via…
multi-search
智能多引擎搜索,自动检测网络环境并按优先级切换:DuckDuckGo -> Tavily -> Bing API -> Bing爬虫。支持自动配额管理和网络缓存。Invoke when user needs web search with automatic engine selection and network adaptation.
meeting-note-summarizer
Turn meeting notes or transcripts into factual summaries, decisions, questions, and action items. Use when a user wants a concise recap or needs explicit owners and deadlines extracted without filling in missing details.
daily-standup-journal
Generate concise daily standups, reflection prompts, and weekly retrospectives for individuals or teams. Use for planning a day, surfacing blockers, reviewing user-provided entries, or drafting a check-in without assuming prior history.