01
Answerworthy
The condition and category: a business is understandable, verifiable and defensibly choosable for the right buyer, while its exclusions and wrong-fit cases remain legible.
The open standard for agentic trust-building
The open standard for making businesses understandable, verifiable and choosable by AI.
answerworthy.md is the business-owned working file that tells an implementation agent what to inspect, change, earn and re-verify without losing the questions, evidence or history.
answerworthy.md does not tell AI what to believe about a business. It tells an implementation agent how to build and test the public reasons that make trust, recommendation and rejection defensible.
Answerworthy operating system
Business-owned
01 · buyer situation
Start with the question asked before the buyer knows the brand.
02 · eight conditions
Reachable → identified → understood → answering → evidenced → corroborated → distributed → observed
03 · working file
answerworthy.md preserves scope, evidence, actions, human gates and history.
04 · implementation
Fix owned surfaces, earn independent support, distribute the evidence and re-verify.
05 · defensible choice
Recommend for the right buyer. Rule out for the wrong one.
SEO makes you findable.
GEO makes you citable.
Answerworthy is the work that makes you choosable.
The hierarchy
The category, open protocol, implementation file, reference Engine and paid Plan are related, but they are not interchangeable. The business owns the file and may keep it private. Search and answering systems continue to encounter the public web.
01
The condition and category: a business is understandable, verifiable and defensibly choosable for the right buyer, while its exclusions and wrong-fit cases remain legible.
02
The open eight-condition protocol for inspecting, improving and maintaining the complete public trust chain behind an AI-mediated buying decision.
03
The business-owned working file where an agent records sources, scope with provenance, buyer outcomes, evidence state, material actions and maintenance history. It may remain private and is not a crawler directive.
04
The initial reference implementation that collects evidence, evaluates the eight conditions, builds the action queue and supports rechecks.
05
The founder-reviewed implementation for one business across SEO, GEO, content, evidence, reviews, PR, authority, partnerships, marketing and distribution.
Why the standard is open
Agents make the full process reachable for small teams and for internal product, marketing, communications and engineering departments. The open standard makes the questions, evidence, decisions and history portable between agents, tools and providers.
Maintenance can recur. Captivity must not.
Monitoring
Asks what an assistant says after the buyer already knows the company name. Useful, but downstream of discovery and choice.
Owned capability
Asks whether the business becomes a defensible option when a buyer describes the problem, constraints or comparison before knowing the brand exists.
Not another crawler directive
Answering systems still retrieve ordinary public surfaces. The standard does not ask them to treat an owner-authored Markdown file as privileged truth.
A business-specific answerworthy.md coordinates the work required to make those surfaces coherent, supportable, accessible and trustworthy.
It is not a replacement for HTML, structured data, robots.txt, sitemaps, llms.txt, public evidence or independent corroboration.
Buyers and answering systems encounter
Implementation agents use answerworthy.md to
A dependency network
The conditions form a dependency network, not a score and not a universal implementation order. Start from the buyer Outcomes and the material blockers in the declared scope.
A tool that checks Reachable and Observed covers two of eight conditions. Record the remaining coverage explicitly; do not infer failure from what was not inspected.
01 · reachable
Can an answering system actually fetch and read the relevant public surfaces?
02 · identified
Can it resolve the business, people, products and aliases without material contradiction?
03 · understood
Can it understand what is offered, for whom, on what terms and where the offer is a poor fit?
04 · answering
Does the public presence answer the pre-brand questions buyers ask before they know the business exists?
05 · evidenced
Can every consequential claim be checked against a method, dated scope, limitations and an inspectable source?
06 · corroborated
Do independent reviews, directories, institutions, experts or customers support the material picture?
07 · distributed
Is the evidence present where relevant buyers, communities, publishers, partners and systems encounter it?
08 · observed
What do assistants actually say for the buyers’ real questions, with provider, locale, date and outcome preserved?
The business-owned working file
Keep answerworthy.md in the repo, shared drive or working folder your agents can access. Publishing it at the website root is optional, not a conformance requirement.
Business claims appear only as objects under evaluation. Inclusion creates no truth or authority.
01
The public or private business material and evidence references needed to continue.
02
Buyer situations, question families, market, alternatives and exclusions, with provenance.
03
The falsifiable buyer conclusions the implementation must make defensible.
04
What was checked across the eight conditions, with exact evidence and uncertainty.
05
The ordered work, human gates, completion tests and later verification.
06
Refresh triggers, re-verification and history-preservation rules.
Put an agent to work
The agent reads the current Standard and inspects any business material already available. If none is available, it asks for one useful source: a website, deck, document, PDF, profile, repository or short description. It discovers Scope before asking further questions.
Get started with https://answerworthy.mdCurrent specification: Answerworthy 1.0-draft · source revision dated 26 August 2026. The agent creates or continues the business file; no completed template is required.
Evidence before persuasion
A bounded metric may help diagnose one question. It never becomes the definition of Answerworthiness or replaces the evidence and work behind it.
The Answerworthy Standard does not collapse a business, market, provider and date into one 0–100 number.
An implementation may calculate a scoped diagnostic when its method, coverage and unavailable inputs remain inspectable.
Every number remains subordinate to the exact question, condition, source, limitation and date behind it.
Changed public surfaces, earned corroboration, explicit gaps and re-verifiable actions matter more than dashboard movement.
verifiedCurrent, inspectable evidence supports the bounded condition inside the stated scope.
unsupportedA consequential claim or proposed answer exists while the evidence required to defend it is absent or insufficient.
contradictedCurrent public evidence materially conflicts with the claim, identity or intended answer.
missingA necessary public surface, answer, proof asset or relationship does not exist.
unreachedThe evidence could not be obtained. No failure, absence or zero is inferred.
staleThe evidence or implementation state is outside its declared freshness boundary.
incompletePart of the condition is established while explicit requirements remain unresolved.
Action states
ready · blocked · waiting-human · waiting-third-party · done
01
Writing a claim into the file gives it no truth, authority or recommendation privilege.
02
Cite what was observed, name what is needed, or preserve both.
03
Blocked, absent or failed observation remains unreached rather than becoming a fabricated negative.
04
Never invent sources, dates, reviews, demand or results, and never rewrite unsupported as established.
05
No fake reviews, astroturfing, self-corroboration or outreach quotas disguised as authority.
06
Agents may draft both; they may assert or send neither without named human authority.
07
Every metric states what it actually observed and never substitutes for the underlying evidence and action.
Keep inputs and material scope premises traceable; unknown and inferred remain explicit.
Connect findings and material next actions to the buyer outcomes they advance.
Read the living Standard, re-verify completed work and preserve useful history.
The implementation loop
01
Check the eight conditions against the real public web and record anything unreachable without guessing.
02
Create or update answerworthy.md with buyer outcomes, scope, exact evidence states and the ordered work.
03
Change only the owned surfaces the principal explicitly authorised, one bounded change at a time.
04
Draft claims and third-party communications, then stop for a human decision before asserting or sending either.
05
Re-fetch every changed surface and compare the result with its falsifiable definition of done.
06
Keep the file, evidence and history so another agent, team or provider can continue without starting over.
The founder-reviewed Answerworthy Plan
The free Read includes the supported Standard result, evidence and material Actions for its declared scope. The paid Plan adds the complete retained research, decision experiments, cross-area dependencies and founder-reviewed priorities for one business, market and language.
The campaign brings prepared marketing drafts and executable briefs across business decisions, evidence production, owned surfaces, independent corroboration, PR, partnerships and distribution. You receive the proposed positioning, copy or pitch where the evidence supports it, with the reason it matters and what happens next.
Where your expertise is needed, we ask precise questions about business facts, evidence or constraints. Each draft identifies those missing inputs and the approval needed before publication or contact. The marketing work arrives prepared for your review.
01
Raw HTML, indexing, architecture, metadata, schema, internal links, performance, accessibility and bot access.
02
Pre-brand situations, question families, answer candidates, comparisons, constraints and observed answer behaviour.
03
Entity resolution, audience, commercial shape, pricing posture, differentiation, fit and explicit wrong-fit cases.
04
Falsifiable qualities, methods, samples, dates, limitations, proof assets and original research.
05
Reviews, directories, marketplaces, customer proof, institutions, experts and correction paths.
06
Editorial narratives, publishers, podcasts, newsletters, events, communities and legitimate earned references.
07
Partnerships, social surfaces, demand creation and placing defensible evidence where it can be encountered.
08
Acceptance tests, indexing checks, answer observations, referrals, regressions and rechecks.
All eight work areas are considered. Only material work enters the campaign; areas needing no change are preserved, and unresolved prerequisites stay explicit.
Why Alexander made it open
Alexander Pichugin created Answerworthy after more than twenty years across web, product, technology, branding and AI, and from frustration with visibility work that repeatedly reports symptoms without leaving the client with an owned operating capability.
Agents now make the complete process accessible to small businesses, lean teams and internal departments. The open standard makes that process portable: another agent, team or provider can continue from the same questions, evidence, decisions and history.
The original provider should remain useful because the work is valuable, not because the client cannot leave.
Canonical open source
The Answerworthy GitHub organisation is the canonical source, history and contribution surface. The public gateway is a category-first projection of that source, not a competing specification.