{ "company": "Firecrawl", "slug": "firecrawl", "website": "https://www.firecrawl.dev", "audit_date": "2026-04-06", "overall_score": 67, "tier": "Emerging", "tier_as_published": "Emerging", "pillars": { "P1": { "name": "Signal Architecture", "score": 20, "max": 25 }, "P2": { "name": "Clarity Stack", "score": 21, "max": 25 }, "P3": { "name": "Trust Envelope", "score": 11, "max": 20 }, "P4": { "name": "Velocity Triggers", "score": 9, "max": 10 }, "P5": { "name": "Gravity Design", "score": 6, "max": 20 } }, "criteria": [ { "id": "P1-A", "pillar": "P1", "name": "Structured Data", "score": 1, "max": 5, "evidence": "No schema.org markup (Organization, Offer, Product) detected on homepage. Website is HTML-based with no JSON-LD schema found." }, { "id": "P1-B", "pillar": "P1", "name": "Machine-Readable Pricing", "score": 5, "max": 5, "evidence": "Pricing is precisely structured: Free (500 credits one-time), Hobby ($16/mo, 3k credits), Standard ($83/mo, 100k credits), Growth ($333/mo, 500k credits), Scale ($599/mo, 1M credits), Enterprise (custom). Credit system is explicit: 1 credit = 1 page. All limits, rates, and billing rules are machine-parseable from https://www.firecrawl.dev/pricing." }, { "id": "P1-C", "pillar": "P1", "name": "llms.txt / Agent Layer", "score": 5, "max": 5, "evidence": "Firecrawl published /llmstxt endpoint as documented service (https://docs.firecrawl.dev/features/alpha/llmstxt). Agent onboarding skill published at https://www.firecrawl.dev/agent-onboarding/SKILL.md. Explicitly designed for AI agent consumption." }, { "id": "P1-D", "pillar": "P1", "name": "API / MCP Availability", "score": 5, "max": 5, "evidence": "Official Firecrawl MCP server published on GitHub (github.com/firecrawl/firecrawl-mcp-server). Clean REST API with three endpoints: Scrape, Search, Interact. Full documentation at https://docs.firecrawl.dev/api-reference/introduction." }, { "id": "P1-E", "pillar": "P1", "name": "Discoverability (GEO)", "score": 4, "max": 5, "evidence": "Trusted by 80,000+ companies. Strong GitHub presence (open source). Product Hunt 5.0 score. High SEO and AI retrieval optimization. Blog posts specifically about agent workflows (e.g., \"Introducing /agent: Gather Data Wherever It Lives on the Web\")." }, { "id": "P2-A", "pillar": "P2", "name": "Offer Completeness", "score": 5, "max": 5, "evidence": "**What:** Web scraping (GET pages as markdown/JSON/HTML/screenshots), search (retrieve full content from web results), crawl (discover all URLs on site), interact (click/type/extract autonomously). **Who:** AI agents, developers, teams. **How Much:** Free $0, Hobby $16, Standard $83, Growth $333, Scale $599, Enterprise custom. All stated on pricing page. Single clear source at https://www.firecrawl.dev/pricing." }, { "id": "P2-B", "pillar": "P2", "name": "Scope & Limits", "score": 5, "max": 5, "evidence": "Every tier has explicit limits: Free 500 credits (one-time), Hobby 3k/month, Standard 100k/month, Growth 500k/month. Concurrent requests: 2, 5, 50, 100 respectively. Rate limits documented. Auto-recharge policy clear. No rollover to next month. All machine-readable." }, { "id": "P2-C", "pillar": "P2", "name": "Substitution Rules", "score": 2, "max": 5, "evidence": "No documented fallback if credits exhausted mid-execution. No overflow handling (auto-upgrade option?). No equivalent-product substitution logic published." }, { "id": "P2-D", "pillar": "P2", "name": "Conditional Logic", "score": 4, "max": 5, "evidence": "SLA conditional: Free/Hobby no SLA. Standard recommended as minimum. Enterprise gets 99.9% SLA. Eligibility gates are explicit and agent-readable." }, { "id": "P2-E", "pillar": "P2", "name": "Semantic Precision", "score": 5, "max": 5, "evidence": "All offer terms precisely defined: \"credit\" = \"1 page scrape\" or \"1 PDF page.\" Concurrent requests = simultaneous API calls allowed. No vague claims (\"powerful,\" \"unlimited*\"). All superlatives backed by specific numbers." }, { "id": "P3-A", "pillar": "P3", "name": "Verifiable Performance", "score": 5, "max": 5, "evidence": "Status page (https://firecrawl.betteruptime.com/) shows 99.99% uptime (as of Feb 4, 2026). Enterprise tier explicitly offers 99.9% SLA. Product Hunt: 5.0 from 10 reviews. Trustpilot: 3.4 from 4 reviews. Third-party verification through status page and review platforms." }, { "id": "P3-B", "pillar": "P3", "name": "Scoped Permissions", "score": 2, "max": 5, "evidence": "No agent-specific permission scoping documented. API key model is all-or-nothing (full access to account's credits and endpoints). No time-bounded, action-bounded, or amount-bounded token options published." }, { "id": "P3-C", "pillar": "P3", "name": "Audit Trail", "score": 1, "max": 5, "evidence": "No machine-accessible audit log mentioned. No transaction history API published for agents to verify past scrape requests and results." }, { "id": "P3-D", "pillar": "P3", "name": "Behavioral Consistency", "score": 3, "max": 5, "evidence": "Status page tracks incidents (last official outage January 21, 2026). Enterprise SLA published. No explicit change policy, version control of terms, or notice period for pricing changes documented." }, { "id": "P4-A", "pillar": "P4", "name": "Friction-Free Activation", "score": 5, "max": 5, "evidence": "**Fully autonomous activation.** 500 free credits issued instantly, no credit card required. API key generated immediately. Agent onboarding skill at https://www.firecrawl.dev/agent-onboarding/SKILL.md enables signup + API key creation in one flow with zero human gates." }, { "id": "P4-B", "pillar": "P4", "name": "Agent Decision Signals", "score": 4, "max": 5, "evidence": "Free tier exists (500 credits). Tier thresholds clear and pricing tier logic explicit. SLA availability gated by tier. Some urgency signals (e.g., /llmstxt endpoint being deprecated June 30, 2025) suggest planned obsolescence triggers, but not comprehensively documented." }, { "id": "P5-A", "pillar": "P5", "name": "Integration Depth", "score": 3, "max": 5, "evidence": "API integration is modular; agents can easily replace Firecrawl with Crawl4AI, ScrapingBee, or other competitors. Content scraped via Firecrawl is immediately portable. Mild switching cost (agent retraining on new API), but not deep application lock-in." }, { "id": "P5-B", "pillar": "P5", "name": "Agent Memory Layer", "score": 1, "max": 5, "evidence": "No agent-accessible memory or context persistence documented. Each scrape request is stateless. No profile, history, or preferences stored for agents across sessions." }, { "id": "P5-C", "pillar": "P5", "name": "Programmatic Renewal", "score": 1, "max": 5, "evidence": "No programmatic renewal API. Credits do not roll over. No agent-accessible usage tracking or re-up signals. Renewal is human-driven (manual upgrade to new tier)." }, { "id": "P5-D", "pillar": "P5", "name": "Compounding Value", "score": 1, "max": 5, "evidence": "Static service offering. No signals showing value improvement with continued use. No training data accumulation, no network effects, no quality metrics improving over time." } ], "strongest_signals": [ { "title": "Agent-Native API & MCP", "detail": "Firecrawl explicitly designed for AI agents. Official MCP server published. /llmstxt endpoint and agent onboarding skill provide agent-native identity layer. This is a reference implementation for agent-ready infrastructure." }, { "title": "Friction-Free Activation", "detail": "Instant API key, no credit card gate, agent onboarding skill enables 1-step signup. Zero human involvement. This is best-in-class for autonomous adoption." }, { "title": "Crystal-Clear Pricing & Limits", "detail": "Every tier, credit cost, and rate limit precisely defined. No ambiguity. Machine-parseable tiers make tier selection trivial for agents." }, { "title": "Verifiable Uptime & SLA", "detail": "99.99% tracked uptime, 99.9% SLA for Enterprise, Product Hunt 5.0. Trust signals are third-party verified." } ], "critical_gaps": [ { "title": "No Agent-Scoped Permissions", "detail": "API keys are all-or-nothing. No way to issue time-bounded, action-bounded, or amount-bounded tokens. This is a security gap for agents executing on behalf of users." }, { "title": "No Audit Trail", "detail": "Agents cannot verify their own past actions. No transaction history or request log accessible to agents." }, { "title": "Gravity Design Missing", "detail": "No renewal signals, no memory layer, no compounding value. Agents have zero incentive to stay with Firecrawl after first use\u2014they can easily switch." }, { "title": "Limited Programmatic Renewal", "detail": "Credits don't auto-renew. No agent-readable usage metrics or renewal eligibility signals." } ], "priority_actions": [], "executive_summary": "Firecrawl scores 67/100 \u2014 **EMERGING** tier. The platform is exceptionally well-built for initial agent adoption: agent-native API, instant activation, crystal-clear pricing, and verifiable uptime. Firecrawl is a reference implementation of agent-ready infrastructure. However, it lacks the deep integrations and trust mechanisms needed for long-term autonomous agent tenancy. Critical gaps: no scoped permissions, no audit logs, no compounding value signals, no programmatic renewal. To reach AGENT-READY (85+), Firecrawl should focus on: (1) agent-scoped token model, (2) machine-accessible audit logs, (3) usage-based renewal signals. These three changes would move Firecrawl from \"easy to start with\" to \"safe to entrust with critical workflows.\"", "rubric_version": "v1-2026-04 (20 criteria, 100 raw points; P3-E Agent Registration added to rubric v2 in 2026-06, not scored in this audit)", "framework": "Agent Native Offers \u2014 The Agent Sale framework", "source_file": "2026-04-06 \u2014 Firecrawl \u2014 Agent Native Offer Audit.md", "rank": 5 }