All 20 Criteria
P1-A Structured Data — 3/5
Homepage has schema.org markup (Organization, WebSite, WebPage schemas with contact details, social profiles on LinkedIn, X, G2, GitHub). Legal entity (Yin Yang, Inc.) and address encoded. No rich Offer/Product/AggregateRating schema found.
P1-B Machine-Readable Pricing — 3/5
Pricing page shows clean HTML tiers ($999/connector/year Expansion; $1,999/connector/year Enterprise) with feature tables. Parseable by scraping but not encoded in schema.org/Offer format or clean JSON.
P1-C llms.txt / Agent Layer — 5/5
robots.txt explicitly references both /llms.txt and /llms-full.txt (per llmstxt.org standard). Both files confirmed present. llms.txt provides structured platform overview; llms-full.txt gives full article content index. Best-in-class signal for agent readability.
P1-D API / MCP Availability — 4/5
"Turn any integration into an MCP server with one API call" — explicitly agent-facing. Dynamic JSON Schema tool generation from API documentation. URL-based cryptographic auth encoding for each MCP server. REST unified API with docs at docs.truto.one. OpenAPI spec not confirmed publicly; no agent card published.
P1-E Discoverability (GEO) — 4/5
llms.txt/llms-full.txt explicitly for LLM consumption. robots.txt allows all crawlers. G2 listing present. Content-rich blog with MCP guides. Strong AI retrieval optimization signals.
P2-A Offer Completeness — 3/5
Pricing page has clear tier names and prices, but the offer is complex: connector-based pricing with minimum launch requirements (10 connectors for Expansion, 60 for Enterprise). What, who, and price are findable but require multiple page reads to fully parse; not in a single machine-parseable source.
P2-B Scope & Limits — 3/5
Some limits explicitly stated: log retention (30-day Expansion / 90-day Enterprise), environment caps (5 / unlimited), and a 14-day trial. API calls described as "unlimited." Rate-limiting handled by RapidBridge internally but no agent-legible rate limit documentation found.
P2-C Substitution Rules — 1/5
No guidance on what happens when an integration is unavailable, degraded, or deprecated. RapidBridge handles rate-limit retries internally but no agent-accessible fallback or substitution rules documented.
P2-D Conditional Logic Transparency — 2/5
Enterprise tier requires sales contact ("Partner with Truto"); terms scattered across pricing, ToS, and DPA pages. No single machine-readable document describing all conditions. Some custom arrangement language implies undisclosed conditions.
P2-E Semantic Precision — 3/5
Mostly precise technical language: "JSONata-based data mapping," "RapidBridge," "SOC 2 Type II." Some vague enterprise marketing ("world-class support," "endless customization"). Functional terminology is accurate and consistent.
P3-A Verifiable Performance Data — 3/5
G2 listing confirmed with positive reviews (responsive support, fast deployment). SOC 2 Type II and ISO 27001 certified. Status page and changelog referenced in footer. No public uptime percentage or third-party SLA dashboard found. Enterprise tier mentions "SLA guarantee" but not publicly disclosed.
P3-B Scoped Permissions — 4/5
Explicit field-level scoping via unified API ("define scopes, limit fields, let customers toggle specific attributes off"). MCP server URL contains cryptographic token encoding which account and which tools to expose — effectively per-agent authorization at the URL layer. Strong agent-specific permission architecture.
P3-C Audit Trail — 2/5
Log retention tiers (30/90 days) documented. Up to 5 environments for isolation. No machine-accessible audit log API for agent systems found; logs appear designed for human review, not programmatic agent consumption.
P3-D Behavioral Consistency — 2/5
Enterprise tier includes SLA guarantee (undisclosed terms). Changelog referenced on site. No version-controlled API terms or explicit notice periods for breaking changes found publicly.
P4-A Friction-Free Activation — 2/5
14-day free trial with credit card required. No instant API key issuance described. Enterprise minimum of 60 connectors requires sales engagement. "Partner with Truto" CTA signals high-touch sales model rather than self-serve activation.
P4-B Agent Decision Signals — 1/5
No programmatic signals for when/why an agent should activate, upgrade, or act. Free trial requires human credit card step. No usage-based triggers or agent-legible entry points published.
P5-A Integration Depth — 4/5
600+ integrations with fully custom JSONata data mappings, on-premises deployment option, dedicated engineers, and custom unified APIs. Rebuilding Truto's configuration layer across 600+ connectors represents extremely high switching cost. Deep lock-in by design.
P5-B Agent Memory Layer — 1/5
No agent-specific memory or personalization layer found. Platform manages credentials and auth state but no per-agent context, history, or preference layer documented.
P5-C Programmatic Renewal — 1/5
Annual billing model ($X/connector/year). No agent-accessible renewal API found. No programmatic mechanism for agents to manage or trigger subscription renewal.
P5-D Compounding Value Signal — 2/5
Value compounds as more connectors are added and custom mappings accumulate. But no agent-visible signal of this compounding value — agents cannot query "what has been built up here" or evaluate switching costs programmatically.
Rubric v1 (April 2026). Scores reflect the company's state on the audit date and may have improved since.