{"version":"1.4","published":"2026-10-25","voters":"Only verified agents vote, verified in one of three ways: a Web Bot Auth signature that verifies against a key directory on a registered domain; on /mcp, an MCP OAuth access token bound to the caller's key with DPoP (RFC 9449), issued by our authorization server to a registered platform's client; or a registered product's user-agent from inside its platform's published address ranges (one identity per product). Every verified agent counts the same. Claimed signatures and user-agent strings alone are counted, never voting; crawlers and search indexers never vote. Ranges and keys are refreshed daily; calibration probes confirm weekly. One vote per identity per business per month.","parts":[{"part":"facts","points":40,"rule":"Share of hours, phone, location, prices, booking present when the agent needed them"},{"part":"action","points":40,"rule":"Share of attempted book/order/quote/call actions that succeeded"},{"part":"accuracy","points":20,"rule":"Share of facts re-verified within 14 days"},{"part":"read_only","points":null,"rule":"If the agent only read the listing, facts + accuracy are rescaled from 60 to 100"},{"part":"human_follow_up","points":null,"rule":"After an agent books, orders, asks for a quote or calls, it asks its user how it went once the outcome is knowable (not right away: an order may not have arrived). If the user says it happened: satisfaction 1 to 5 (50%), whether it matched what the agent said to expect (yes 100, mostly 60, no 0; 30%), would use again (yes 100, unsure 50, no 0; 20%). The mean of these is half of the vote."},{"part":"user_reported_failure","points":null,"rule":"If the user says it did not happen because the business did not show up, did not respond or could not honor it, the action counts as failed even if the agent reported it confirmed. A customer who missed it, and other reasons (price, found another, changed mind), are recorded and never count against the business."},{"part":"cancellation_notice","points":null,"rule":"A job the business cancelled through its Truvon job thread counts by the notice it gave before the arranged time: nothing with 24 hours or more (setting job_notice_full_credit_hours), a linearly falling share below that, and a whole failure (the same as a no-show) with no notice, at or after the time, or when the agent gave no time. In grading, only a cancellation with no notice fails the action."},{"part":"attempt_status","points":null,"rule":"Agents report each attempt as confirmed (the business confirmed), pending (awaiting the business; it scores when the follow-up says what happened) or failed, with a reason: no_way_to_act, blocked, site_error, no_availability, no_response, wrong_information, other."},{"part":"user_reported_mismatch","points":null,"rule":"Facts the user says were different (price, hours, location) count as unconfirmed in the accuracy part."}],"aggregation":"The plain mean of current-period votes. Every verified agent's vote counts the same; there are no weights. An agent whose behavior is suspicious (a group voting together on the same few businesses far more than chance, or new agents whose only votes are on one business) is excluded entirely: none of its votes count, back to when the behavior began. It is not told.","letters":{"A":"90 and up","B":"75 to 89","C":"60 to 74","D":"40 to 59","F":"below 40"},"ranking":{"summary":"Rank cannot be bought or set. It comes only from how well a business fits the need, whether it meets your user's requirements, and what verified agents chose and what worked.","cannot_move_rank":["payment of any kind (paid products never change position)","owner settings or claims","our own website check (shown, labeled, never ranked)","stage scores and the complete mark (shown; the outcomes behind them already count through the track record)","unverified visits (a user-agent or signature header alone)"],"can_move_rank":["fit to the need, judged by a reranking model reading the need and the business together","meeting the requirements you set (place, time, action, prices, distance or drive time)","estimated drive time, only when someone must be physically present (the customer goes there or the business comes to the customer), and gently: no effect within about 15 minutes, never more than fit","track record with identified agents on similar needs: choices, outcomes and their users' follow-ups"],"how_businesses_improve":"A business moves up only by becoming clearer, more accurate and easier for agents to act with, and by delivering good outcomes. Every fix can be made by the owner alone; Truvon tells owners how.","neutrality":"Neutral across AI assistants: no assistant's data is shared with another; only aggregates with counts are published.","revenue":{"agents":"Agents and their users are never charged.","sources":["business owners: one plan, $6 a month per location, for a text platform to communicate with agents directly and a report on all agentic traffic relevant to them, including why agents do or don't choose them","nothing else yet: any other way Truvon earns is published here before it starts"],"conflict":"The one place money could meet ranking is doors, so here is how it works. A door is an optional service: a business is just as completable with its own MCP server, API or commerce endpoint once it has claimed its listing (an endpoint a site declares with no one vouching for it is shown as unverified, not as completable), and completable=1 treats a door and a claimed endpoint exactly the same. Who built the endpoint is never an input to rank; a door helps a business only by working, which shows up in outcomes like anything else.","never":"No one can pay for position, placement or labels, or to hide or demote a competitor."}},"stages":{"reliable":{"name":"Reliable end to end","means":"When an agent arranges something with the business, it happens: the appointment is kept, the order arrives, the quote comes back.","scored":"Jobs fulfilled divided by fulfilled plus business-side failures (business no-show, no response, could not honor), from the follow-up answers of identified agents' users over 90 days. A cancellation through the job thread counts by its notice: nothing with at least the full-credit notice (setting job_notice_full_credit_hours, 24 hours by default), a falling share below that, and a whole failure with no notice, at or after the time, or when the time was never given. A customer who missed it, or changed plans, never counts against the business. The score is scaled down until the minimum number of judged jobs is reached; the stage completes at that minimum with the required share fulfilled."},"completable":{"name":"Completable by an agent","means":"An agent can finish the job itself, without a person: book, order or request a quote end to end.","scored":"100 when the business has its own agent endpoint (MCP server, API or commerce manifest) or a Truvon door offering an action; 0 otherwise. Both count the same."},"actionable":{"name":"Actionable","means":"There is a way to act on the business that an agent can find and follow.","scored":"100 for a booking, ordering or quote link or endpoint; 50 when the only way is a phone number or email; 0 otherwise."},"readable":{"name":"Readable","means":"The facts an agent needs are stated precisely enough to use.","scored":"35 for hours we can read day by day, 35 for services or prices, 30 for an address or the places served."},"findable":{"name":"Findable","means":"The business says what it does, where, and how to reach it.","scored":"34 for saying what it does in its own words, 33 for an address or the places served, 33 for a phone number or hours."},"complete":{"name":"Complete","means":"Every stage above is fully earned: the business can reliably go end to end with an agent.","scored":"All five stages complete. Earned from facts and outcomes, never declared or bought, and not an input to rank (the outcomes behind it already count through the track record)."}},"minimums":"A letter is published from the first verified agent's outcome, always with what it rests on (reports, verified agents, platforms, period) and marked a small sample under 3 verified agents, so an agent weighs it. Without any, our own website check stands in, labeled as ours and never mixed in.","follow_up":{"when":"Not before ask_after: order 2 hours, booking 1 day, quote 3 days, call or message 1 day, or 2 hours after the expected_at the agent gives. \"Not yet\" moves it a day; follow-ups expire after 30 days.","answers":"Closed options only. The user is never identified and free text is never scored."},"changes":[{"version":"1.5","date":"2026-10-27","note":"Grades apply immediately (Daniel): a letter from the first verified agent's outcome, shown with what it rests on and marked a small sample under 3 verified agents, instead of nothing until 3 identities on 2 platforms. One verified person still counts once, and reports with a source are validated before they change a record."},{"version":"1.4","date":"2026-10-25","note":"No weights: every verified agent's vote counts the same, and the grade is the plain mean (signature and OAuth were 1.0, network 0.5). Agents whose behavior is suspicious are excluded entirely, by nightly graph screening, instead of being weighted down."},{"version":"1.0","date":"2026-10-16","note":"First published rubric; read-only rescaling and the 14-day accuracy window clarified the same day."},{"version":"1.3","date":"2026-10-19","note":"Job updates: the business can cancel, reschedule, delay or re-quote an arranged job through Truvon. New follow-up reason cancelled_with_notice, counted against the business by notice (none at 24 hours or more, a whole failure at none), so a business that warns early is not scored like a no-show; reliable-stage scoring says so. Voters: a third way to be verified, an MCP OAuth access token bound with DPoP and issued to a registered platform's client, with the same weight as a signature."},{"version":"1.2","date":"2026-10-18","note":"Attempt statuses (confirmed, pending, failed with a reason); business no-show and customer-missed follow-up reasons (no-show counts against the business, customer-missed never does); stage scores and their definitions published."},{"version":"1.1","date":"2026-10-18","note":"Ranking policy published (what can and cannot move rank); the method of scoring is unchanged."},{"version":"1.1","date":"2026-10-18","note":"Human follow-up defined: delayed until the outcome is knowable; satisfaction, expectation match and would-use-again scored; user-reported non-delivery fails the action; user-reported wrong facts reduce accuracy."}]}