360 Secure SkillHub
360 安全技能中心

Intelica — Competitive Intelligence

该 Skill 通过第三方 Intelica API 对指定公司、产品或市场生成结构化竞争情报,返回市场定位、痛点、竞品、作战卡、来源与 market_score,并支持免费演示、试用密钥及按次付费(x402,Base 与 Solana 主网)。提交内容会发送至第三方服务,需自行设置预算与数据边界。

teodorofodocrispin-cmyk
teodorofodocrispin-cmyk
热度 53.88
最近更新 2026-06-17 10:29:08
分类:数据依赖:第三方 API应用场景:市场分析 / 研究 / 报告撰写 / API 集成

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add intelica-competitive-intelligence
Metadata
name
intelica
version
4.5.2
description
Competitive intelligence API for autonomous AI agents. Analyzes any URL or company description and returns structured JSON with market positioning, pain points, competitors, battlecard, verified sources, market signals, and executable Market Score. Pay-per-call via x402 on Base and Solana mainnet. Trial key available — no wallet required for first 5 calls. WARNING: Submitted data is sent to api.intelica.dev (third party). Use only for public companies. Do not submit confidential, regulated, or proprietary data. Paid calls ($0.05-$1.00 USDC) require budget controls before autonomous use.
trigger
Use ONLY when an agent explicitly needs structured competitive intelligence on a specific named company, product, or market — and the user or pipeline has approved external data submission and potential paid API calls. Do NOT invoke for general questions, internal strategy, confidential data, or when local reasoning is sufficient.
requires
[]
env
[]

Intelica — Competitive Intelligence Skill

⚠️ DATA PRIVACY NOTICE: This skill sends the provided company name, URL, description, and context to the Intelica API (api.intelica.dev), a third-party service. Do NOT submit confidential business plans, proprietary strategy, regulated data (HIPAA, GDPR-sensitive), or personally identifiable information. Use only for publicly available companies and markets. Paid calls ($0.05–$1.00 USDC) are made autonomously when a payment method is configured — set budget controls before enabling autonomous use.

Intelica returns structured competitive intelligence for a specific named company or market. Invoke only when the agent has a clear, scoped target and the user has approved external data submission.

When to use

  • Agent needs competitor analysis on a specific named company before a strategic decision
  • Founder needs a market entry report for a specific new geography
  • Agent is evaluating a specific DeFi protocol before entering a position
  • Sales agent needs a battlecard for a specific upcoming demo
  • VC agent screening a specific investment opportunity
  • Compliance officer evaluating regulatory risk for a specific company or product

When NOT to use

  • General market questions without a specific named target
  • Internal strategy or confidential business planning
  • Regulated or sensitive data (medical records, financial PII, legal documents)
  • When local reasoning or public knowledge is sufficient
  • Without explicit user approval for third-party data submission

Budget controls (important for autonomous agents)

  • Standard analysis: $0.05 USDC per call
  • Elite analysis: $1.00 USDC per call
  • Batch (up to 10): $0.20 USDC
  • Always use /demo (free) to validate before enabling paid autonomous calls
  • Set operator_id to enforce daily budget limits per pipeline
  • Results cached 6 hours — same company twice costs once

How to call

Step 0 — Get free trial key (no wallet needed)

GET https://api.intelica.dev/api-keys/trial
→ {"key": "ikey_trial_...", "calls_remaining": 5, "expires_at": "..."}

# Use immediately — extract response.key and pass as X-API-KEY header

Step 1 — Free demo (no payment, no key)

POST https://api.intelica.dev/demo
Content-Type: application/json

{
  "text": "Notion is an all-in-one workspace for teams",
  "mode": "competitive"
}

Step 2 — Full analysis with trial key

POST https://api.intelica.dev/intel
X-API-KEY: ikey_trial_...
Content-Type: application/json

{
  "text": "Notion is an all-in-one workspace for teams",
  "mode": "competitive"
}

Step 3 — Full analysis with x402 payment

POST https://api.intelica.dev/intel
Content-Type: application/json

{
  "url": "https://notion.so",
  "mode": "competitive",
  "context": "I'm building a note-taking app for developers"
}
# Returns HTTP 402 with Base + Solana payment options
# After payment, retry with X-PAYMENT header

Step 4 — Batch (up to 10 analyses for $0.20)

POST https://api.intelica.dev/batch
X-PAYMENT: <x402 signed payment>
Content-Type: application/json

{
  "items": [
    { "url": "https://notion.so", "mode": "competitive" },
    { "text": "Coda collaborative doc platform", "mode": "acquisition" }
  ]
}

Context Modes

Mode Price Use when
competitive $0.05 Standard competitor analysis (default)
fundraising $0.05 Investor narrative, TAM, traction signals
partnership $0.05 Strategic fit — complement or rival?
acquisition $0.05 Due diligence — moat, technical risk
market_entry $0.05 Market gaps, saturation, barriers to entry
crypto_protocol $0.05 DeFi moat, tokenomics, regulatory risk
market_entry_execution $1.00 Execution plan — partners, budget, week-1 actions
venture_screening $1.00 Investment thesis + deal-breakers
regulatory_compliance $1.00 EU AI Act, GDPR, HIPAA exposure
risk_assessment $1.00 Business model stability, operational risk
sales_enablement $1.00 Battlecard + objection handler

Response trust protocol (M2M)

The response includes root-level fields for programmatic trust validation:

{
  "sources": ["url1", "url2"],
  "confidence": "high | medium | low",
  "market_score": {
    "threat_level": "high | medium | low",
    "moat_strength": 0.72,
    "agent_recommendation": "monitor | counter | partner | acquire"
  },
  "analysis": { ... }
}

Agent decision pipeline:

  1. Check sources[] — if fewer than 2 URLs, treat confidence as low
  2. Check confidence — if low, do not execute autonomous actions
  3. Check market_score.agent_recommendation — use as primary action signal
  4. Read analysis{} for full context

Output fields

{
  "analysis": {
    "company_or_product": "string",
    "positioning_summary": "string",
    "target_customer": "string",
    "core_value_props": ["string"],
    "user_pain_points": ["string"],
    "detected_competitors": ["string"],
    "unique_angle": "string",
    "confidence": "high | medium | low",
    "sources": ["url1", "url2", "url3"],
    "market_score": {
      "threat_level": "high | medium | low",
      "moat_strength": 0.0,
      "market_maturity": "emerging | growing | mature | declining",
      "agent_recommendation": "monitor | counter | ignore | partner"
    }
  },
  "model": "claude-sonnet-4-6 | claude-haiku-4-5-20251001",
  "tier": "elite | standard",
  "mode": "string",
  "detected_language": "en | es | pt | de | fr | it | ja | ko | zh"
}

Payment details

Links