SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillspm-competitor

pm-competitor

Writing & research

View repository

Updated 10-16 06:22

About this skill

Preamble (run first)

# 检查前置
if [ -f "docs/01-需求调研/市场调研报告.md" ]; then
  echo "✅ 市场调研报告 - 已找到"
else
  echo "⏳ 市场调研报告 - 未找到（可选，缺失时由本技能建立基线）"
fi
if [ -f "docs/01-需求调研/竞品监控月报.md" ]; then
  echo "🔁 竞品监控月报 - 已存在（将基于上期做本期对比）"
else
  echo "🆕 竞品监控月报 - 首次建立基线"
fi

Write-time boundaries

When only discussing, do not write files; startup checks do not create directories. When the user asks to save artifacts, first confirm the current product, the exact output path, and whether to overwrite or append to existing files; after confirmation,only immediately before the actual Writecreate the parent directory of that output file. For full installation, see skills/_shared/artifact-policy.md; installing this skill alone also follows this paragraph.

Cross-Agent interaction rules

When the process requires interaction with the user:

  1. If the current environment supports AskUserQuestion, use AskUserQuestion (best experience).
  2. If the current environment does not support AskUserQuestion, you must ask the same question using a normal chat message.
  3. Ask only one question at a time.
  4. After asking, you must stop the current turn and wait for the user to answer (STOP and WAIT).
  5. Do not generate documents or write to docs before the user answers.
  6. Existing docs files cannot replace the user's answer in this round.

Applicable scenarios

  • The user says "monitor competitors", "competitor intelligence", "how are competitors changing", "competitor monthly report", "competitor change alerts"
  • Distinguish from pm-market / pm-search : those areone-time research; this skill isperiodic monitoring + change alerts, emphasizing baselines and comparisons.

Execution flow

Step 1: Determine monitoring targets and dimensions (main agent - user interaction)

Use AskUserQuestion to ask:

🎯 Competitor monitoring setup

Which competitors do you want to monitor? Which dimensions do you want to monitor?

A) Automatically use the competitor list from the market research report B) I will manually specify competitors (please provide names) C) Monitor only the top 1-3 core competitors

Monitoring dimensions (multiple selections allowed): version/feature updates, pricing changes, market/funding actions, user public sentiment, hiring/organizational signals

Record to variable COMPETITORS and WATCH_DIMS。

Step 2: Establish/read the monitoring baseline (main agent)

If docs/01-需求调研/竞品监控月报.md does not exist → enterfirst baseline mode：

Use the Agent tool to dispatch subagents in parallel to take a full snapshot of each competitor (search first):

Tool: Task
Parameters:
  subagent_type: "general-purpose"
  description: "竞品基线调研-{竞品名}"
  prompt: |
    你是竞品情报分析师。请对竞品「{竞品名}」建立当前快照。

    **搜索优先级**：首选 AnySearch（定位脚本后 python3 调用），失败降级 Exa MCP，最后 WebSearch（标注「⚠️ 降级模式」）。
    **监控维度**：{WATCH_DIMS}
    **要求**：提取最新版本/功能、定价、近期动态、舆情基调（正面/中性/负面）、关键信号。
    **输出 JSON**：
    ```json
    {
      "competitor": "{竞品名}",
      "snapshot": {"version": "...", "pricing": "...", "news": "...", "sentiment": "正面/中性/负面", "signals": ["..."]},
      "sources": [{"title": "...", "url": "..."}]
    }
    ```

If a previous monthly report already exists → read it and extract the previous baseline values for comparison in this period.

Step 3: Set the monitoring cadence and alert thresholds (main agent - user interaction)

Use AskUserQuestion to confirm:

📅 Monitoring cadence

A) Monthly monitoring (recommended) B) Biweekly monitoring C) Quarterly monitoring

Change alert thresholds (examples):

  • Pricing change ≥ {X}%
  • Major version/core feature launch
  • Concentrated outbreak of negative public sentiment
  • Major funding/organizational changes

Record to variable CADENCE and ALERT_RULES。

Step 4: Dispatch subagents to collect intelligence for this period (core optimization)

Dispatch subagents in the same way as in Step 2 (in parallel), but add to the prompt "compare with the previous period and mark changes":

对比上期基线：{上期快照}
重点识别本期相对上期的异动（新增/变更/消失），并判断影响（高/中/低）。

The main agent collects all competitor results.

Step 5: Generate monthly competitor monitoring report + anomaly alerts (main agent)

Use the Write tool to generate docs/01-需求调研/竞品监控月报.md：

# {产品名称} 竞品监控月报（{周期}）

## 文档信息
- 周期: {起} ~ {止}
- 监控竞品: {列表}
- 生成工具: super-pm / pm-competitor

---

## 一、本期异动预警（重点）
| 竞品 | 异动 | 类型 | 影响 | 我方建议 |
|------|------|------|------|---------|
| {竞品} | {异动} | 版本/定价/舆情/融资 | 高/中/低 | {建议} |

## 二、竞品快照对比
| 维度 | 上期 | 本期 | 变化 |
|------|------|------|------|
| {竞品A·版本} | ... | ... | ... |
| {竞品A·定价} | ... | ... | ... |
| {竞品B·舆情} | ... | ... | ... |

## 三、行业信号
- {融资/并购/新政等}

## 四、我方机会与威胁
- 机会: {从竞品弱点中找}
- 威胁: {从竞品动作中找}

## 五、下一步建议
1. /pm-position - 基于竞品动态重新校准定位
2. /pm-priority - 将竞品威胁纳入需求优先级
3. /pm-market - 需要深度专项研究时

Step 6: Recommend next steps (main agent)

✅ Monthly competitor monitoring report generated:docs/01-需求调研/竞品监控月报.md

Recommended actions:

  1. /pm-position - Calibrate product positioning
  2. /pm-priority - Convert competitor threats into requirements
  3. /pm-market - In-depth specialized research

Search dependency notes

  • AnySearch(preferred) → Exa MCP(alternative) → WebSearch(fallback, marked "⚠️ degraded mode").
  • For sentiment-type monitoring, prioritize vertical domains (tech media, app store reviews, social platforms).

Output quality comparison

✅ Good example：

  • Has baseline comparison: "Competitor A priced at ¥99 last month, raised to ¥129 this month (+30%), triggering an alert"
  • Has impact assessment and recommendation: "Impact: high; Recommendation: evaluate whether we should raise prices too or emphasize cost-effectiveness"
  • Has source: "Source: Competitor official website announcement 2026-07-20"

❌ Bad example：

  • Lists news without comparison: "Competitor A released a new feature"
  • No impact assessment
  • No source

Common pitfalls / Red Flags — STOP

Pitfall Correct approach
Treating monitoring as a one-off study Monitoring must have a baseline + periodic comparison
Piling up news without alerts Must mark anomalies and impact levels
No source Attach a source to each anomaly
Ignoring recommendations for our own actions The endpoint of monitoring is recommendations for our strategy

Output quality check / Verification Checklist

  • Monitoring targets and dimensions determined
  • Previous baseline established or read
  • Cadence and alert thresholds set
  • Anomaly alerts + snapshot comparison produced
  • Output document generated to docs/01-需求调研/
  • 2-3 follow-up skills recommended

⚠️ If any item fails → complete it before marking as done.