Skip to Content
AI AgentsAgent SkillsSkill Reference

Agent Skills Reference

How to read this page

Every skill depends on the CompetLab MCP server being configured — that’s how it reaches your projects, competitors, and dimensions (see Setup). Beyond that, each entry notes anything extra it uses: live web research, public third-party APIs, or one of the free scan tools. Skills trigger on plain-language requests; the Ask examples are phrasings each skill advertises, not fixed commands.

Orchestrator

competlab-cmo-report

Runs most of the suite to compose a CMO report for a project — a live, on-demand composition. It sequences the five dimension dashboards and most of the other skills — the analysis and research skills, though not the sales battlecard or the weekly delta briefing — and fetches the platform’s monthly Strategic Briefing through get_briefing, then synthesizes a layered set of documents: one main report, a dozen per-dimension docs, and a deep dive per Tier-1 competitor. The CMO report is composed live from the skills; the Strategic Briefing it pulls in is the platform’s persistent, longitudinal product — the skill fetches that, it doesn’t produce it. This is the one-command “give me the whole picture” skill.

  • Uses: a broad set of the mcp__competlab__* tools (the dashboards, the Strategic Briefing, competitors, and URL fetch), plus the sub-skills it runs, web research, and URL verification.
  • Ask: “CMO report for Acme” · “full competitive analysis for my project” · “cold-start briefing”.
  • Effort: high — needs the other skills installed, since it calls them.

Lagging-indicator skills

These five work from the data CompetLab already monitors — what’s measured, scored, and tracked over time. They’re the original set, and each runs standalone.

competlab-ai-visibility

How ChatGPT, Claude, and Gemini mention and recommend your brand versus competitors, built from your AI Visibility Score data — the latest scores, per-provider breakdowns, and trends. For understanding your standing in AI answers, not traditional SEO.

  • Uses: the AI Visibility dashboard, trend, history, and check-detail tools; web research.
  • Ask: “AI visibility report” · “how do LLMs see my brand” · “what does ChatGPT say about us” · “GEO analysis”.

competlab-weekly-briefing

An on-demand delta briefing — what changed across your monitored dimensions, what it means, and what to do — that you run whenever you want a pulse. Short and focused on the delta since your last look. Distinct from CompetLab’s monthly Strategic Briefing.

  • Uses: all five dimension dashboards, the Strategic Briefing, alerts, and the content changelog.
  • Ask: “weekly briefing” · “competitive update” · “what’s new with competitors”.

competlab-competitor-dive

A comprehensive single-competitor dossier combining your monitoring data with live web research: tech stack, pricing, positioning, content strategy, AI visibility, public sentiment, a SWOT, and recommended responses.

  • Uses: all five dashboards plus positioning/pricing history and competitor detail; web research.
  • Ask: “deep dive on Rival Inc” · “competitor dossier for rival.com” · “SWOT for Rival Inc”.

competlab-battlecard

A sales-ready battlecard for one competitor — a quick-reference card for calls, demos, and objection handling, built from monitoring data and live research.

  • Uses: the five dashboards and competitor data; web research.
  • Ask: “battlecard for Rival Inc” · “how to beat Rival Inc” · “objection handling for Rival Inc”.

competlab-landscape

A full competitive landscape across all five monitored dimensions: market dynamics, competitive matrices, cross-dimensional patterns, and strategic recommendations. For the whole field, not a single competitor.

  • Uses: all five dashboards, the Strategic Briefing, schedules, and AI Visibility trend; deep web research.
  • Ask: “competitive landscape” · “full CI analysis” · “quarterly competitive review”.
  • Effort: high

Leading-indicator skills

These seven research live public signals that tend to move before the monitored dimensions — funding before press, hiring before launches, MCP presence before category-wide agent adoption. They lean on public third-party sources and web research, and verify every URL before it lands in a deliverable.

competlab-status-watch

Probes each competitor’s public status page and incident history — ongoing outages, incident frequency, post-mortem quality. A missing or broken status page is itself a signal.

  • Uses: list_competitors, the fetch_url scan tool, and direct probes of common status-page backends.
  • Ask: “competitor reliability” · “status page check” · “operational maturity”.

competlab-funding-watch

Surfaces recent funding, ownership changes, ARR estimates, M&A, and executive transitions, adapting its read to each vendor’s funding mode (public, PE-owned, bootstrapped, VC-stage, or M&A-volatile).

  • Uses: competitor list, fetch_url for verification, Perplexity, and web research.
  • Ask: “funding posture” · “competitor funding” · “ARR estimates”.
  • Effort: medium

competlab-ai-ecosystem

Measures external developer-ecosystem signals around each competitor: GitHub organization and top repos, npm/PyPI download volumes, community-built (not first-party) MCP servers, integration-marketplace presence. Complements agent-adoption, which measures first-party signals.

  • Uses: public GitHub, npm, and PyPI APIs (a GH_TOKEN lifts GitHub’s rate limit); web research and Perplexity.
  • Ask: “developer ecosystem for Rival Inc” · “who’s building on competitor X”.

competlab-hiring-signals

Probes hiring and go-to-market motion through public ATS APIs (Ashby, Greenhouse, Lever, Workable), with a LinkedIn fallback and research on executive transitions. Most reliable for Series-B-and-later and public vendors.

  • Uses: public ATS APIs via shell, Perplexity, and URL verification.
  • Ask: “competitor hiring” · “is Rival Inc expanding its sales motion” · “exec transitions”.
  • Effort: medium

competlab-agent-adoption

Reports each competitor’s Agent Adoption posture. It wraps CompetLab’s Agent Adoption scan — discoverability, access control, content readability, agent endpoints — and adds protocol-level verification of any MCP-server URLs the scan finds, because a browser returning 200 doesn’t mean a working MCP server exists.

  • Uses: the agent-adoption scan tools, check_ai_crawlers, fetch_url, and direct protocol probes.
  • Ask: “does Rival Inc have a real MCP server” · “Agent Adoption posture in my category”.

competlab-product-watch

A snapshot of a competitor’s product surface beyond the homepage: structured changelogs, GitHub Releases, named-asset directories, API-doc versions, deprecation patterns. Snapshot-only — it doesn’t compute velocity, diffs, or alerts.

  • Uses: the content changelog, public product surfaces via shell, and fetch_url.
  • Ask: “what did Rival Inc ship recently” · “product launches in my category”.

competlab-customer-voice-snapshot

A snapshot of competitor presence on review platforms (G2, Capterra, Trustpilot) and recent complaint themes. Snapshot-only with limited coverage by design, since review platforms broadly block automated access — it leans on grounded research with URL verification for citations.

  • Uses: competitor list, Perplexity, and URL verification.
  • Ask: “what do customers say about Rival Inc” · “G2 reviews for Rival Inc”.

The full suite

Orchestrator
  competlab-cmo-report          composes a live CMO report (runs most of the suite)

Lagging-indicator (from monitoring data)
  competlab-ai-visibility       AI Visibility Score report across ChatGPT, Claude, Gemini
  competlab-weekly-briefing     on-demand delta briefing
  competlab-competitor-dive     single-competitor dossier + SWOT
  competlab-battlecard          sales battlecard
  competlab-landscape           full five-dimension landscape

Leading-indicator (live public signals)
  competlab-status-watch        reliability / incident history
  competlab-funding-watch       funding, ownership, ARR, M&A, exec moves
  competlab-ai-ecosystem        GitHub / npm / PyPI / community MCP presence
  competlab-hiring-signals      hiring + GTM motion via public ATS
  competlab-agent-adoption      Agent Adoption posture + MCP verification
  competlab-product-watch       product-surface snapshot
  competlab-customer-voice-snapshot   review-platform snapshot

The data these skills read comes through the MCP server — see its tool reference for the underlying tools, and Setup to install the plugin.

Last updated on
Setup