Methodology
How agenticweb.wiki is built, what counts as a row, how state is assessed, and where the gaps are.
The nine steps
The map is organised around nine steps an agent and a site pass through, each with a classic-web analog.
| # | Step | Question | Analog | Definition |
|---|
| 1 | Discover | What is here, and where? | sitemap, .well-known | How an agent learns that a resource, tool or agent exists at an origin, and where to find its description. |
| 2 | Identify | Who is this agent? Who is this site? | TLS certificate, WHOIS | How an agent proves who it is to a site, and how a site or another agent proves who it is back. |
| 3 | Permit | What may the agent do here? | robots.txt, terms of use | How a site states what agents may read, train on, or do, and how agents obtain authorisation to act. |
| 4 | Read | Give me the content | HTML, RSS | Representations of a site's content made for machine readers rather than browsers. |
| 5 | Act | Do something for me | forms, REST APIs | How an agent invokes a capability a site or tool exposes. |
| 6 | Coordinate | Talk to another agent | none; agent-to-agent is new | How agents delegate to, negotiate with, and message other agents. |
| 7 | Pay | Settle the bill | HTTP 402, finally used | How an agent pays for a resource, authorises a payment on a person's behalf, or completes a purchase. |
| 8 | Present | Show a human the result | the browser | How an agent renders output or a live interaction to a person. |
| 9 | Account | Prove what happened; be judged | logs, audit trails | How actions are recorded, verified after the fact, attributed and measured. |
How records are produced
Records are drafted by AI research agents that read each primary source — spec text, datatracker page, or vendor documentation — under a fixed brief published at /methodology/verify-brief/. Every evidence label is assigned per source, not per record: a claim is marked VERIFIED only when an agent read a primary source or two independent secondaries itself. Every URL in the data is fetched, and its response is recorded in the linkcheck. An adversarial review re-fetches a sample of sources before publish and corrects or downgrades anything it cannot confirm.
Inclusion
A row needs a public specification text or a fixed artifact; products without a spec are not rows.
Facet vocabularies
Parties
| id | label | definition |
|---|
| agent-site | agent ↔ site | An agent interacting with a website or API origin. |
| agent-agent | agent ↔ agent | Two agents interacting directly. |
| agent-human | agent ↔ human | An agent presenting to, or taking instruction from, a person. |
| agent-payment | agent ↔ payment network | An agent and a payment rail, wallet, or card network. |
| agent-registry | agent ↔ registry | An agent and a third-party index or registry. |
Form
| id | label | definition |
|---|
| well-known-file | well-known file | A fixed-path file under /.well-known/ or the site root. |
| http-header | HTTP header | A request or response header. |
| endpoint | endpoint | An HTTP or RPC endpoint with a defined message format. |
| dns-record | DNS record | A record type or name convention in DNS. |
| onchain-registry | on-chain registry | A smart contract acting as a registry. |
| document-format | document format | A file format or document convention. |
| browser-api | browser API | A JavaScript API available to pages in a browser. |
Track
| id | label | definition |
|---|
| ietf | IETF | Internet Engineering Task Force draft or RFC. |
| w3c | W3C | W3C working or community group output. |
| linux-foundation | Linux Foundation | A Linux Foundation project or sub-foundation. |
| industry-body | industry body | A trade association such as IAB Tech Lab. |
| vendor-consortium | vendor consortium | Several vendors publishing jointly without a standards body. |
| single-vendor | single vendor | One company controls the specification. |
| community | community | Open community project with no formal governance. |
| academic | academic | Research output without a standards process. |
Scope
| id | label | definition |
|---|
| general | general | Applies to any site or agent. |
| commerce | commerce | Buying and selling goods or services. |
| advertising | advertising | Ad buying, measurement, or attribution. |
| payments | payments | Moving money. |
| coding-tools | coding tools | Developer tooling and editors. |
Phase
| id | label | definition |
|---|
| idea | idea | Proposed, no specification text. |
| draft | draft | Specification text exists and is changing. |
| working-group | working group | Adopted by a standards working group. |
| review | review | In formal review or last call. |
| candidate | candidate | Feature-complete, seeking implementations. |
| stable | stable | Published as stable or final. |
| living | living | Continuously updated with no versioned finals. |
| deprecated | deprecated | Superseded or discouraged by its issuer. |
| withdrawn | withdrawn | Withdrawn or archived. |
| unknown | unknown | Not established. |
Implementation
| id | label | definition |
|---|
| none-found | none found | No implementation located. |
| prototype | prototype | Reference or experimental implementation only. |
| developer-trial | developer trial | Behind a flag or origin trial. |
| shipping | shipping | At least one production implementation. |
| multi-implementation | multi-implementation | Independent production implementations from more than one party. |
| deprecated | deprecated | Implementations being removed. |
| unknown | unknown | Not established. |
Adoption
| id | label | definition |
|---|
| unmeasured | unmeasured | Nobody has published a count. |
| claimed | claimed | A party asserts adoption; the claimant is named. |
| observed | observed | We or a third party have seen instances but not counted systematically. |
| measured | measured | Counted by a documented probe run. |
| unknown | unknown | Not established. |
Evidence
| id | label | definition |
|---|
| VERIFIED | verified | Primary source, or two independent secondary sources, checked by us. |
| REPORTED | reported | One secondary source. |
| THIN | thin | Existence established; detail not. |
Freshness
Each row is re-checked on a 30-day cadence. Freshness state:
| id | label | definition |
|---|
| current | current | Verified within the review cadence. |
| review-due | review due | Past cadence; not yet re-checked. |
| stale | stale | Past twice the cadence. |
| source-unavailable | source unavailable | Primary source no longer resolves. |
Blanks
Steps with three or fewer standards found as of 2026-09-07. The gap is the finding, not an oversight.
Named gaps: no published standard was found for agent attention or ranking signals, for agent-side audit logs beyond attestation proposals, or for agent-to-human consent receipts. These are recorded as blanks, not omissions.
What the occupants list
Checked 2026-09-06. For each reference-map occupant we track, rows we have that they do not, and rows they have that we do not.
agentswelcome
| We have, they don't | They have, we don't |
|---|
- agent-client-protocol
- ap2-checkout-mandate
- l402
- erc-8183
- erc-8226
- anthropic-agent-skills-format
- agents-md
- aipref-content-usage-attachment
- aipref-vocabulary
- tdmrep
- web-bot-auth-signature-agent-card
- ard
- ai-catalog-well-known-file
- ans
- aid
- ardp
- ads
- a2ui
- ag-ui
- osi
- okf
- openai-apps-sdk-manifests
- aamp
- artf
- adcp
- vi
- spiffe
- wimse
- aauth
- daap
- aap
- aip
- agtp
- vcap
- agora-protocol
- agora-commerce-protocol
- s402
- agent-attribution-protocol
- agent-audit-trail
- api-catalog-rfc9727
- agent-permissions-json
- auth-md
- oauth-metadata-for-agents
- ai-crawler-tokens-robots
- mcp-apps
- vercel-agent-readability
- ai-txt
|
|
Listed by them, excluded here (not a specification)
- Agentic AI Foundation (AAIF)
- World Wide Web Consortium (W3C)
- Cloudflare Pay Per Crawl (Pay Per Crawl)
- Schema.org structured data vocabulary (for agents) (schema.org)
- AGNTCY (Internet of Agents project) (AGNTCY)
lumar
| We have, they don't | They have, we don't |
|---|
- agent-client-protocol
- l402
- erc-8004
- erc-8183
- erc-8226
- anthropic-agent-skills-format
- agents-md
- aipref-vocabulary
- web-bot-auth-signature-agent-card
- http-message-signatures
- ard
- ai-catalog-well-known-file
- dns-aid
- ans
- aid
- adp
- ardp
- ads
- oasf
- a2ui
- ag-ui
- osi
- okf
- openai-apps-sdk-manifests
- aamp
- artf
- adcp
- tap
- vi
- spiffe
- wimse
- aauth
- daap
- aap
- aip
- agtp
- vcap
- anp
- agora-protocol
- agora-commerce-protocol
- s402
- agent-attribution-protocol
- agent-audit-trail
- slim
- api-catalog-rfc9727
- rsl
- agent-permissions-json
- oauth-metadata-for-agents
- mcp-apps
- vercel-agent-readability
- agents-json
- agents-txt
|
|
Listed by them, excluded here (not a specification)
- Canonical URL
- Editorial metadata
- HTTP Link header
- JSON-LD
- Language and encoding
- Markdown twin
- OpenAPI description
- Product / Offer schema
- Server-side rendering
- XML sitemap
- schema:Action
agent-ready
| We have, they don't | They have, we don't |
|---|
- agent-client-protocol
- ap2-checkout-mandate
- l402
- erc-8004
- erc-8183
- erc-8226
- webmcp
- agents-md
- aipref-content-usage-attachment
- aipref-vocabulary
- tdmrep
- cloudflare-content-signals-policy
- http-message-signatures
- ard
- ai-catalog-well-known-file
- dns-aid
- ans
- aid
- ardp
- ads
- oasf
- ag-ui
- osi
- okf
- openai-apps-sdk-manifests
- aamp
- artf
- adcp
- tap
- vi
- spiffe
- wimse
- aauth
- daap
- aap
- aip
- agtp
- vcap
- anp
- agora-protocol
- agora-commerce-protocol
- s402
- agent-attribution-protocol
- agent-audit-trail
- slim
- rsl
- auth-md
- ai-crawler-tokens-robots
- mcp-apps
- agents-txt
- ai-txt
|
|
Listed by them, excluded here (not a specification)
- Content parity (anti-cloaking)
- Endpoint reachability (RFC 9309)
- WCAG 2.2 + layout stability
What answer engines read
As of September 2026, no answer engine — Google AI Overviews or AI Mode, ChatGPT Search, Perplexity, Claude, or Bing Copilot — documents consuming llms.txt, ARD catalogs, Agentmap, AIPREF, or TDMRep. Google's own AI optimisation guide states that no special markup or llms.txt is required for a page to appear in AI features. This site serves those files because it maps them, not because they affect ranking. Source: developers.google.com/search/docs/fundamentals/ai-optimization-guide.
Probes
Where a standard defines a checkable request, the protocol row carries an executable probe: method, path, what to parse, and what to record when the path is absent. Adoption counts on this site are marked measured only when we have run that probe ourselves; otherwise adoption is claimed, observed, or unmeasured.
