跳至正文
OpenAgentSkill
浏览已开放 · 目录记录日期 2026-11-21 · 自动更新和账户服务暂时关闭。
其他技能

trade-show-finder

Compare and prioritize trade shows using current official sources, ICP fit, goals, and execution readiness.

LensmorOfficial · 已有静态检查记录

安装与使用

运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。

npx skills add LensmorOfficial/trade-show-skills --skill trade-show-finder

技能说明


name: trade-show-finder description: Compare and prioritize trade shows using current official sources, ICP fit, goals, and execution readiness. license: MIT metadata: version: 0.5.0 stage: pre-show category: research homepage: https://github.com/LensmorOfficial/trade-show-skills/tree/main/trade-show-finder

Trade Show Finder

Help B2B exhibitor teams decide which shows deserve budget, team time, and follow-up.

When this skill triggers:

  • Read references/show-fit-framework.md before scoring or recommending
  • For shortlist discovery, comparison, or annual planning, also read references/show-archetypes.md
  • Treat this as a show selection skill, not a generic event-directory lookup

Example Requests

  • Which trade shows should we go to?
  • 哪些展会值得参加
  • Welche Messen lohnen sich?
  • どの展示会に出展すべき?
  • ¿A qué ferias asistir?

Workflow

Step 1: Determine Request Mode

Use one of these four modes:

  1. Specific-show decision Example: "Should we exhibit at MEDICA 2026?" Default outcome: Exhibit, Attend only, or Skip

  2. Named-show comparison Example: "Compare Interpack and PACK EXPO for us" Default outcome: side-by-side winner with tradeoffs

  3. Shortlist discovery Example: "Find the best packaging shows in Europe for a mid-market automation vendor" Default outcome: ranked shortlist with scores

  4. Annual planning Example: "What 3 shows should we prioritize this year?" Default outcome: top priorities by tier, not an exhaustive directory dump

If the user is only asking for a factual lookup ("When is MEDICA 2026?"), answer the fact directly, then offer a one-line follow-up such as "If you want, I can score whether it's worth exhibiting for your ICP."

Step 2: Collect Decision Inputs

For comparison, discovery, and annual planning, prioritize these business inputs:

  • What the company sells
  • ICP / target company type
  • Buyer titles or functions
  • Primary goal: pipeline, distributor search, partnerships, brand visibility, launch, or market entry
  • Target region(s)
  • Whether the team plans to exhibit or only attend

Optional inputs:

  • Budget band
  • Team size
  • Timeframe
  • Deal size / revenue target

Rules:

  • Ask only for missing decision-critical inputs
  • Do not fall back to generic questionnaires
  • If the show is already named, do not ask for industry or region just to restate the obvious
  • If the year is ambiguous for a named show, ask which edition and stop. Never infer the edition from the current date, repository context, or model memory.

Step 3: Build a Curated Candidate Set

Do not behave like a fresh web crawl every time.

For discovery, comparison, and annual planning:

  • Start from the candidate seeds and archetypes in references/show-archetypes.md
  • Narrow the set based on vertical, buyer, region, and go-to-market goal
  • Keep user-named shows in the set even if they score poorly

For every show you keep:

  • Verify dates, venue, website, and recent scale with web search
  • Prefer official sites for current-edition facts
  • Use directories or third-party roundups only as backfill
  • If a site errors or is blocked after 1-2 tries, move on and mark the uncertain field as est. or TBC

Evidence gate before scoring:

  • Confirm the intended edition on an official event or organizer page.
  • Confirm at least one current-edition official source for audience, exhibitor categories, program focus, or other evidence directly relevant to the user's ICP and goal.
  • A remembered reputation, city, event name, or prior-edition fact is not enough to pass this gate.
  • If the gate cannot be satisfied, stop with Verification required. Do not produce a Show Fit Score or an Exhibit / Attend only / Skip decision. Return the missing official-source checks instead.
  • In a Verification required response, do not repeat remembered reputation, city, month, scale, audience, or buyer-profile claims; do not give an "offline proxy" or any directional recommendation. Restate only user-supplied facts, the exact evidence missing, and the next official-source lookup actions.
  • Do not guess an organizer domain. Link it only after the URL has been found and verified.
  • Never invent alternative events. A suggested alternative must have a real, verified official website.

Collect, when available:

  • Official show name
  • Dates
  • City and venue
  • Official website
  • Exhibitor count
  • Visitor count
  • Core buyer or attendee profile
  • Product / category fit
  • Frequency

Prioritize usefulness over exhaustiveness. If a show is clearly weak for the user's ICP or objective, drop it rather than padding the list.

Step 4: Score the Shows

Use the scoring method in references/show-fit-framework.md.

For every serious recommendation, provide:

  • Show Fit Score (0-100)
  • Execution Readiness: Ready, Conditional, or Not assessed
  • Recommendation band
  • Decision: Exhibit, Attend only, or Skip
  • A short Why not line that surfaces tradeoffs

Use these recommendation bands:

  • 80-100: Priority 1 — exhibit
  • 65-79: Priority 2 — exhibit if budget permits, or attend first
  • <65: lower priority — attend only or skip

If budget band, team size, or travel complexity are missing, set Execution Readiness to Not assessed rather than guessing.

Step 5: Write the Response

Every substantial response should use this structure:

## Executive Recommendation
[One-paragraph answer with the top decision]

## ICP / Goal Snapshot
- Company / offer:
- ICP:
- Buyers:
- Goal:
- Region:
- Motion: Exhibit / Attend

## Shortlist or Comparison Table
| Show | Dates | Location | Show Fit Score | Decision | Why it fits |
|------|-------|----------|----------------|----------|-------------|

## Show Fit Score
[Brief score explanation by dimension]

## Evidence Used
- [Offici
阅读完整来源 ↗