Public draft · 0.1.0-draft.1

The open contract for declaration, evidence, action and recheck.

The Answerworthy Standard defines a public owner manifest and a portable evidence contract for independently evaluating and improving answerworthiness.

A PDF may explain the result. The canonical bundle preserves what was observed, what was inferred, what remained unavailable, what should happen next and who is allowed to do it.

Read the raw specification Verify provenance

Version: 0.1.0-draft.1
Status: Public draft
Published: 5 August 2026
Specification: CC BY 4.0
Tooling: Apache-2.0

0. Abstract

A declaration can be portable without pretending to verify itself.

The public /answerworthy.md manifest declares identity, scoped claims, sources, answerable questions, known corrections, coverage limits and freshness policy in a format people can edit and machines can parse.

Its existence proves control of the publishing origin. It does not establish independent truth, importance, citation eligibility or endorsement by an answer engine.

Conforming implementations may collect evidence into answerworthy.bundle.json, derive a permission-aware plan as answerworthy.plan.md and publish machine-readable results in answerworthy.conformance.json.

Human reports, including PDF, are optional projections of the same canonical record.

1. Artifact architecture

Four artifacts. Four different authorities.

The standard separates what the subject says, what an implementation observed, what should happen and what conformance tests actually passed.

Answerworthy artifact chain0.1

01/answerworthy.mdThe owner-published identity, scoped claims, sources, answerable questions, corrections, limits and freshness policy.02answerworthy.bundle.jsonThe dated evidence, provider states, findings, uncertainty, scores and provenance that an independent implementation actually obtained.03answerworthy.plan.mdA permission-aware action graph with dependencies, evidence requirements, completion tests, verification and recheck instructions.04answerworthy.conformance.jsonThe exact standard version, profiles, tests, failures, withheld claims and certification state for one bundle.

2. Core invariants

The standard refuses the shortcuts that make AI-readiness scores impressive and untrustworthy.

I-01

No evidence, no evaluated claim

A pass, warning or failure references retained evidence. Missing evidence becomes unavailable, manual or not applicable.

I-02

Unknown is not zero

Blocked access, provider failure, parser failure and absent imports never become fabricated absence or zero visibility.

I-03

Declaration is not verification

A public manifest proves control of the publishing origin. It does not prove truth, independence or likely citation.

I-04

One source of truth

The JSON bundle is canonical. Markdown, HTML and PDF are projections and cannot strengthen its certainty.

I-05

Side effects denied by default

An action record grants no credentials, publication authority, outreach permission or third-party consent.

I-06

No guaranteed outcome

The standard does not guarantee rankings, citations, recommendations, traffic, leads or revenue.

3. Canonical dimensions

Answerworthiness is bounded by subject, buyer question, language, market, provider and time.

Every finding maps to one canonical dimension or a namespaced extension. An implementation may score the dimensions but must publish coverage, method, caps and withheld states.

access-retrieval

Access & retrieval

Can relevant systems retrieve and render the necessary public evidence?

entity-certainty

Entity certainty

Can the subject be identified without material ambiguity or contradiction?

offer-comprehension

Offer comprehension

Can a qualified buyer or agent understand what is offered, for whom and under which conditions?

evidence-corroboration

Evidence & corroboration

Can consequential claims be supported, attributed and independently checked?

buyer-question-coverage

Buyer-question coverage

Does the public evidence answer priority pre-brand questions, comparisons and constraints?

answer-performance

Answer performance

What was actually mentioned, cited, recommended or followed through in declared answer-system observations?

4. Agentic action graph

A task is not agentic until authority, evidence and completion are explicit.

Every action identifies its source findings, targets, dependencies, owner, evidence requirements, guardrails, definition of done, verification and rollback or irreversibility.

advisory

May be read, compared and planned. It grants no execution authority.

human_required

Requires a named human decision or performance.

agent_ready

May be executed only inside separately granted credentials and bounded scope.

external_authorisation_required

Depends on permission from a third party, platform, publisher or other external actor.

Action contract

action_id: action-publish-methodology
permission_mode: human_required
source_findings:
  - finding-methodology-unavailable
dependencies:
  - action-confirm-public-claims
definition_of_done: >
  A public methodology page explains scope, evidence classes,
  unavailable states and recheck rules.
verification:
  - Fetch the public HTML and confirm the bounded claims.
  - Re-run the relevant criteria.

5. Conformance profiles

Profiles describe evidence actually obtained, not code that merely exists.

core

Portable subject, run, evidence, finding, provider-state and limitation contract.

agentic

Permission-aware action graph and answerworthy.plan.md projection.

observed-web

Dated browser, search or external-public-source observations.

answer-probe

Provider, model, prompt and replicate observations with four outcomes kept separate.

outside-buyer

Validity-gated comprehension, trust, relevance, recommendation and next-action evidence.

recheck

Baseline comparison separating implemented, unresolved, regressed, new and unverifiable states.

integrity

SHA-256 digests for the canonical bundle and every declared projection.

Self-declared conformance ships its record. “Answerworthy Certified” remains reserved for a separately published certification programme.

6. Discovery and interoperability

Designed to sit beside the web stack, not impersonate it.

Draft 0.1 uses /answerworthy.md as the normative manifest path. /.well-known/answerworthy.md is an experimental mirror until a well-known URI registration exists.

The manifest may be linked from llms.txt, sitemaps and ordinary pages. A standards-compliant discovery hint may use rel="describedby"; the draft does not claim a registered answerworthy link relation.

HTML discovery

<link
  rel="describedby"
  type="text/markdown"
  href="/answerworthy.md"
/>

The standard does not replace robots.txt, sitemaps, JSON-LD, ordinary indexing requirements or provider-specific controls. It is not presented as an IETF, IANA, W3C or search-engine-endorsed standard.

7. Raw normative artifacts

Read, parse and implement the draft.

Specification.md ↓Conformance.md ↓Bundle schema ↓Conformance schema ↓