← Back to Skills Marketplace
gomisterb

agentconcierge

by GoMisterB · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
44
Downloads
1
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install agentconcierge
Description
Find and recommend AI agents for any workflow using AgentConcierge. Use when someone asks which AI tools or agents would help them for their role, task, or p...
README (SKILL.md)

AgentConcierge

Discover the best AI agents for any workflow by querying AgentConcierge — a catalog of 9,800+ agents across OpenAI, Anthropic, n8n, HuggingFace, Zapier, and more. Returns 5 personalized recommendations ranked by match score.

When to Use

  • "What AI agents should I use for [task]?"
  • "Find me the best tools for [role] who needs help with [pain point]"
  • "I'm a [job title] — what agents would save me the most time?"
  • "Recommend AI tools with a [$X] budget using [tools]"
  • "What's the best agent for email, content, data analysis, scheduling, etc.?"

When NOT to Use

  • You already know the specific agent name and just want info (search the web instead)
  • Non-AI tools or general software outside the AI agent space
  • Requests unrelated to discovering or comparing AI agents

Required Context

Before calling the API, collect from the conversation:

Field Required Example
role Yes "Sales / Business Development"
painPoint Yes "Cold outreach and follow-up cadences"
tools No "HubSpot, Slack, Google Workspace"
budget No "$50" or "$200" or "$500+"
teamSize No "solo" or "small team"

If role or painPoint are missing, ask one concise follow-up before calling the API.

API Call

curl -s -X POST https://agentconcierge.io/api/recommend \
  -H "Content-Type: application/json" \
  -d '{
    "role": "Sales / Business Development",
    "painPoint": "Cold outreach and follow-up cadences",
    "tools": "HubSpot, Slack",
    "budget": "$50",
    "teamSize": "small team"
  }'

Response shape:

{
  "recommendations": [
    {
      "name": "Instantly AI",
      "tagline": "Automates cold email sequences at scale",
      "category": "automation",
      "pricingModel": "subscription",
      "pricePerMonth": 37,
      "tags": ["email", "outreach", "sales"],
      "url": "https://instantly.ai",
      "matchScore": 94,
      "matchReasons": [
        "Built for SDRs doing high-volume cold outreach",
        "Native HubSpot integration to sync contacts automatically"
      ]
    }
  ]
}

Presenting Results

Format each recommendation clearly:

🤖 Found 5 AI agents matched to your profile:

1. **Instantly AI** — 94% match
   Automates cold email sequences and follow-ups at scale.
   ✓ Built for SDRs doing high-volume cold outreach
   ✓ Native HubSpot integration to sync contacts automatically
   💰 $37/mo — https://instantly.ai

2. **Apollo.io** — 91% match
   ...

After presenting, offer:

"Want to save these to your stack at agentconcierge.io/dashboard?"

Error Handling

Status Action
429 Tell user the API is rate-limited; retry in 60 seconds
400 Missing role or painPoint — ask user for the missing field
500/503 Suggest visiting https://agentconcierge.io/search directly

Notes

  • Rate limit: 10 requests per 60 seconds per IP
  • Budget $0 or "free only" returns only free/freemium agents
  • Same role + different pain point = different agent set
  • Full catalog browsing: https://agentconcierge.io/search
  • See references/api-schema.md for complete request/response schema
Usage Guidance
This skill appears coherent and low-risk, but note it sends whatever you provide (role, pain point, tools, budget) to https://agentconcierge.io. Do not include sensitive or proprietary information (passwords, internal secrets, private customer data, or detailed confidential business data) in queries. The registry metadata lists no separate homepage even though files reference agentconcierge.io — if you're concerned, verify the service/owner (agentconcierge.io) and its privacy policy before sending real data. If you want to test first, try a few non-sensitive sample queries to confirm responses and rate-limit behavior (10 req / 60s) before using it with real workflows.
Capability Analysis
Type: OpenClaw Skill Name: agentconcierge Version: 1.0.0 The agentconcierge skill is a legitimate tool designed to recommend AI agents based on user-provided roles and pain points. It interacts with a public API (agentconcierge.io) using standard curl commands and does not attempt to access sensitive local files, execute arbitrary code, or exfiltrate unauthorized data. All instructions in SKILL.md and documentation are aligned with its stated purpose.
Capability Assessment
Purpose & Capability
Name/description promise (find and recommend AI agents) aligns with the runtime instructions: collect role/painPoint/etc. and POST them to https://agentconcierge.io/api/recommend using curl. Required binary (curl) and the single brew install are proportional and expected.
Instruction Scope
SKILL.md only instructs the agent to collect a small set of user-provided fields (role, painPoint, optionally tools/budget/teamSize) and send them to the stated third-party API. It does not instruct reading local files, environment variables, or other system state. Note: this does transmit user-supplied conversational data to an external service (agentconcierge.io), which is expected behavior but worth considering for sensitive inputs.
Install Mechanism
Install spec is a single brew entry (curl). This is low-risk and proportionate; no downloads from untrusted URLs or archive extraction are present. The skill is instruction-only (no code written to disk by the skill itself).
Credentials
No environment variables, credentials, or config paths are requested. The skill does not require secrets or cloud credentials, which is appropriate for a public recommendation API.
Persistence & Privilege
always is false and the skill is user-invocable; it does not request permanent presence or modify other skills. No elevated privileges or cross-skill configuration changes are requested.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install agentconcierge
  3. After installation, invoke the skill by name or use /agentconcierge
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Find and recommend AI agents for any workflow using AgentConcierge. Use when someone asks which AI tools or agents would help them for their role, task, or pain point. Searches 9,800+ agents across all major marketplaces and returns 5 ranked recommendations with match scores and reasons.
Metadata
Slug agentconcierge
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is agentconcierge?

Find and recommend AI agents for any workflow using AgentConcierge. Use when someone asks which AI tools or agents would help them for their role, task, or p... It is an AI Agent Skill for Claude Code / OpenClaw, with 44 downloads so far.

How do I install agentconcierge?

Run "/install agentconcierge" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is agentconcierge free?

Yes, agentconcierge is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does agentconcierge support?

agentconcierge is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created agentconcierge?

It is built and maintained by GoMisterB (@gomisterb); the current version is v1.0.0.

💬 Comments