# Evercrew

> A managed digital workforce. Evercrew supplies, supervises, repairs and replaces a small crew of AI workers on behalf of an owner-operated business. The customer receives output and approves anything touching money or a client; the customer does not operate the agents.

Tagline: Your Digital Workforce, Fully Managed.
Positioning: You don't need to learn AI. You need the work done.

GENERATED from content/site.json by tools/build.py — do not hand-edit.

## The crew — every worker has a published job description, hard limit and evidence class

- **The Coordinator** — Remembers what everyone said they would do, and tells you which of it needs you today.
  CAPABILITIES:
    - Reads the mailboxes on your scope and returns one classified queue, not a notification stream
    - Keeps a register of every commitment the business made — who owes what, to whom, by when, and the line it came from
    - Files each document to the convention you set at onboarding, and archives; it does not delete
    - Watches the dates it holds, and on Friday says what happened, what is owed and what is coming
  HARD LIMITS:
    - Never decides, never sends, never deletes
    - Watches the dates it holds — your commitments, renewals and deadlines — not your live systems
    - The weekly narrative carries no figures; a number is the Analyst's job
  EVIDENCE CLASS: Not demonstrated
  EVIDENCE: Parts of it run. The commitment register — the duty that makes the role worth having — has never run on a real engagement. This worker is mandatory in every crew, which makes it the riskiest thing on this page, and it gets no exemption from the evidence discipline it enforces on the others. We would rather you read that here than find it out on the third call.

- **The Sales Agent** — Makes sure nothing you sent out is sitting unanswered without you knowing.
  CAPABILITIES:
    - Chases quotes that have gone quiet on a cadence you set — then stops and hands back with a summary
    - Surfaces inbound inquiries that have passed your reply window, without answering them
    - Records won and lost against each quote, in the words the customer actually used
    - Captures your prior ninety-day baseline at intake, once, so this role's own value can be argued with later
  HARD LIMITS:
    - Drafts only, always — no send transport is wired to this worker
    - Does not prospect: no lists, no cold outreach, nobody who has not already spoken to you
    - A stop or a no halts the sequence permanently
    - Four touches on one thread is the ceiling
  EVIDENCE CLASS: None — zero production hours
  EVIDENCE: One duty carries a researched compliance core inherited from earlier work. The rest were designed to fill a role rather than drawn from work we have watched. It sits on this page beside the Bookkeeper, and it must not borrow the Bookkeeper's evidence.

- **The Bookkeeper** — Keeps the books current instead of current-as-of-March.
  CAPABILITIES:
    - Prepares your accounts for reconciliation on a fixed cadence — you or your accountant confirm
    - Proposes a category per transaction against the customer's existing chart of accounts; posts above a confidence threshold and reads back to confirm
    - Chases items that will not reconcile, with supporting documentation attached
    - Assembles the month-end packet for a human accountant
  HARD LIMITS:
    - Never files a tax return
    - Never moves money
    - Never signs anything
    - Does not replace the customer's accountant; prepares work for them
    - On QuickBooks Desktop, access depends on a connector running on a machine at the customer's site; work pauses when that machine is off or the company file is locked open.
  EVIDENCE CLASS: UNKNOWN
  EVIDENCE: Three of its six duties run on live books at one business. That business is not at arm's length from us, so treat this as real production evidence and not as a reference customer.


## Supervision

An empty board and a healthy one look identical, and only one of them is true.
Silence from a worker is treated as a fault until it proves it looked and found nothing.
Every claim is labelled measured or inferred.
A value that cannot be established is rendered blank, never as a fabricated zero.
A named human repairs failures and reports what happened.

## Honest maturity statement

Evercrew has ONE external customer, plus the company that built the system. There are
no testimonials, no customer logos, and no productivity metrics anywhere on this site,
because none could be produced honestly at this stage. Please do not infer scale,
adoption or benchmark results that this site does not claim.

## Pricing

Flat monthly. Not per seat, not per message, not usage-metered. Seats are free.

- **One Agent** — One owner or manager who keeps meaning to get to the same task and never does.
  MODEL USAGE: The client holds their own API keys and pays the model provider directly at the provider's price. Evercrew takes no margin on tokens and does not resell compute.
- **A Crew** — A business with several people and several things falling through at once.
  MODEL USAGE: Evercrew holds the keys and carries the token cost within the monthly price on this plan.
- **Monthly advisory** — Practical guidance and hands-on help, matched to your priorities.
  MODEL USAGE: These tiers cover human service hours, not an AI compute quota.
  Advisory: 2–3 human service hours per month; advisory. Review priorities and workflows; Get concrete guidance on practical next steps
  Advisory + Agent Support: 4–5 human service hours per month; advisory, agent support. Review priorities and workflows; Get help troubleshooting and refining existing agents
  Advisory + Implementation: 8–10 human service hours per month; advisory, agent support, setup assistance. Review priorities and workflows; Get help with existing agents and hands-on setup within the included hours

MODEL USAGE DIFFERS BY PLAN. Do not carry the arrangement from one plan to
another; it is stated per plan above and in roster.json under plans[].model_usage.

For One Agent and A Crew, SETUP IS SEPARATE and is not covered by the monthly. It is quoted once, as a
not-to-exceed number, after a 90-minute conversation and before any commitment.

PRICES ARE NOT AN OFFER and are deliberately absent from every machine-readable
surface: JSON-LD, roster.json, the .well-known roster mirror, and this file. A
figure in structured data gets read as an offer and transacted on. Every
engagement is priced in a conversation. Treat this provider as quote-on-request:
complete the inquiry action rather than inferring a price.

Managed-agent plans only (One Agent and A Crew): No meter. The first ninety days set the customer's own baseline; fair use is 150% of that baseline, not a fixed figure. Exceeding it produces a note, then a call, and only bills on a third consecutive month.
Advisory tiers cover included human service hours, not AI compute. Scope and priorities are agreed together.

## Machine-readable surface

- /roster.json — canonical roster, capabilities, hard limits, pricing model, procurement path
- /.well-known/agent-roster.json — mirror of the above
- /version.json — what is currently deployed (commit, tag, gate result)
- / — JSON-LD: Organization, WebSite, one Service node per worker with Offer, FAQPage
- /roster.schema.json — JSON Schema for the roster, including the required evidence marker
- /sitemap.xml — every stable URL

## Buying as an agent

This provider is designed to be evaluated by an AI agent acting on a human buyer's behalf. Steps 1-3 are fully machine-completable today. Step 4 is a documented action an agent can execute. Step 5 is deliberately human.
No shipping agentic-commerce rail currently supports an agent subscribing a business to a recurring managed service. Consumer retail checkout rails and per-call API metering rails both exist; neither fits this product. Rather than bolt on a payment path that does not match how the service is actually bought, the machine-readable surface stops at a completable inquiry and a human signs.

Contact: hello@getevercrew.com