AI Competitor Intelligence Agent

September 10, 2026 · View on GitHub

An AI agent for competitive intelligence — cross-cutting audits combining SEO, ads, social, review, and pricing data into one competitor-vs-client comparison — backed by real data APIs.

Part of Agency Agents OS, an open ecosystem of specialized AI agents for real business work.

Agency Agents OS — the central catalog this repo is part of.
ai-seo-agent — supplies the live domain/backlink data this repo's competitive-audit sub-agent cross-cuts.
ai-ads-agent — shares this repo's ad-library-mining capability.
ai-social-agent — feeds the social-presence side of a competitive audit.
ai-reputation-agent — feeds the review/sentiment side of a competitive audit.
MuAPI MCP docs — connect this repo's SKILL.md files via MCP.
MuAPI access keys — create the API key this agent needs.

What this covers

This repo is the umbrella for anything an agency or in-house team would call "the competitor intelligence agent": pulling a competitor's SEO position, ad activity, social presence, and review/reputation signals into a single comparison report against the client, rather than checking each channel by hand. It deliberately cross-cuts other umbrellas — ai-seo-agent, ai-ads-agent, ai-social-agent, ai-reputation-agent — pulling one signal from each into a combined view instead of duplicating any of their individual capabilities.

Sub-agents

Agent	Does	Status
Competitive Audit	Cross-cutting audit of a competitor combining SEO position, social presence, and review/reputation signals into one comparison report against the client	Blueprint
Ad Library Mining	Analyzes which ads a competitor is actively running — creative, messaging, and longevity as a signal of what's working	Blueprint
Price & Trial Monitoring	Extracts a competitor's public pricing and free-trial details from their own site, and diffs against an earlier snapshot to catch changes	Blueprint

Required Muapi APIs

seo.search_performance — competitor and client organic search position/ranking signals (see ai-seo-agent).
ads.ad_library_search — public ad-library-style lookup of a competitor's actively running ads (see ai-ads-agent).
social.read_posts — competitor and client social account activity and engagement (see ai-social-agent).
reputation.review_search — competitor and client review/reputation signals across public review sources (see ai-reputation-agent).
company.products — extracts a competitor's own pricing/plan data by domain or direct URL. Owned by this repo.

These capabilities are shared with the sibling umbrella repos above; this repo's sub-agents combine them into cross-channel comparison reports rather than owning any one channel.

See each sub-agent's SKILL.md for the specific capabilities it uses.

Setup

Create a Muapi account and API key at muapi.ai.
Review the Muapi API quickstart and OpenAPI schema for the relevant endpoints.
Load the SKILL.md for the sub-agent you need into your agent runtime (hosted agent, MCP client, or custom LLM app), or follow it manually.

Using with an AI agent

Every sub-agent's SKILL.md is model- and runtime-agnostic — it's plain Markdown, so it works with any LLM agent, not just Claude. Two integration paths:

As an MCP connection (the agent gets live Muapi tools):

Muapi runs an MCP server at https://api.muapi.ai/mcp that any MCP-compatible client can connect to — Cursor, Windsurf, Claude, or your own custom agent.

Cursor / Windsurf / other clients with a header field: connect to https://api.muapi.ai/mcp with an Authorization: Bearer YOUR_MUAPI_KEY header.
claude.ai / Claude Cowork / other connector UIs with no header field: use the URL-embedded key form instead, https://api.muapi.ai/mcp/YOUR_MUAPI_KEY, via Settings → Connectors → Add custom connector.
Claude Code / Claude Desktop: claude mcp add muapi -e MUAPI_API_KEY=YOUR_MUAPI_KEY -- muapi mcp serve (uses the muapi CLI's stdio transport — Claude Code's HTTP MCP client doesn't reliably inject tools).

Full setup details for every client: muapi.ai/docs/mcp.

As agent instructions (any LLM follows the workflow directly):

Drop a sub-agent's SKILL.md into a Claude Code project's .claude/skills/ directory, paste it into a custom-GPT/Project's system instructions, hand it to an autonomous agent framework as a tool spec, or attach it directly in a chat conversation — then ask the agent to follow it.

Read-only vs. write actions

All sub-agents are read-only — they gather and compare public signals into a report. Neither publishes, files a complaint, responds to a review, or takes any action on a competitor's or client's own accounts.

Status and limitations

Ad Library Mining is Blueprint: it only needs ads.ad_library_search, which is coded but not yet callable end-to-end (the aggregator it routes through needs a vendor API key that hasn't been provisioned) — see its own SKILL.md for exact scope (currently-visible listings only, no spend/impression data). Competitive Audit moved Coming Soon → Blueprint (2026-09-09): its SEO signal (seo-domain-overview/seo-backlinks-history) and social signal (social.read_posts, all platforms except linkedin) are both live and tested; its reputation signal (seo-business-reviews) is live but Google-only — Amazon/app-store/Trustpilot review data is not yet wired up. The sub-agent runs a real, partial comparison today rather than waiting on 100% channel coverage, with the reputation gap flagged explicitly in its completeness note. Price & Trial Monitoring is Blueprint (added 2026-09-10): it only needs company.products, which is coded but not yet callable end-to-end (same aggregator/vendor-key blocker as Ad Library Mining) — see its own SKILL.md for exact scope (one-shot snapshot per call, diffed manually against a prior snapshot; no built-in history or scheduling).

Contributing

See Agency Agents OS CONTRIBUTING.md.

License

MIT