# Witnessability

> Witnessability examines how evidence and explicit reasoning support a particular claim, under stated conditions and limits. This public research site provides explanations, versioned source links, and the preserved Witnessability Conceptual Core 1.0 publication.

The guides are editorial explanations, not amendments to versioned publications. WCC 1.0 is the stable Conceptual Core release (version 1.0.0, tag v1.0.0); Witnessability Model 1.1 is a separate published research paper. WEXP, CBEE and ATR references are individual Internet-Drafts, not RFCs or Internet Standards. Follow each publication's exact version, status and qualifications.

Use the original Core Markdown and signed release records for questions about the released text. A matching hash or valid signature establishes a bounded integrity result; it does not independently establish the signer's identity, governance authority, conceptual truth or complete coverage. Licenses apply to the components identified in the signed License Grant Record, not automatically to the entire website or these discovery resources.

The links below provide public reading resources. No account or API token is needed to read them. Markdown alternatives retain the canonical HTML page as their citation target; the original Core download remains the versioned source.

The site's robots.txt declares search=yes and ai-input=yes for public resources served here; reading responses repeat these preferences. AI training is unspecified by Content Signals. Existing licenses, attribution requirements and third-party rights still apply; these preferences do not relicense artifacts or grant rights in externally linked publications.

Automated readers should send a descriptive User-Agent identifying their own application. The hosting platform may reject generic Python-urllib identifiers with HTTP 403 / error 1010. A browser identity is not required, and User-Agent text does not authenticate a client.

## Start here

- [Site overview](https://witnessability.io/index.md): Purpose, entry points and the relationship between evidence, claims and verification.
- [What is witnessability?](https://witnessability.io/what-is-witnessability/index.md): A short introduction to assessable claims, grounds and limits.
- [Model and Conceptual Core](https://witnessability.io/model/index.md): Distinguish conceptual vocabulary, evidence-and-argument structure and evaluation profiles.
- [Guide directory](https://witnessability.io/guides/index.md): Choose a practical question and follow its detailed explanation.

## Applied questions and vocabulary

- [Audit trail completeness](https://witnessability.io/guides/audit-trail-completeness/index.md): Conditions needed before missing records can support an absence claim.
- [Verification, appraisal and reliance](https://witnessability.io/guides/verification-appraisal-and-reliance/index.md): Separate artifact checks, evidentiary findings and decisions to act.
- [Tool calls versus execution](https://witnessability.io/guides/tool-call-vs-execution/index.md): Distinguish a request, recorded execution and an observed external effect.
- [Observability versus witnessability](https://witnessability.io/comparisons/observability-vs-witnessability/index.md): Diagnostic telemetry and support for a specific proposition have different questions and limits.
- [Worked examples](https://witnessability.io/examples/index.md): Requests, archival preservation and missing records, with bounded conclusions.
- [Glossary](https://witnessability.io/glossary/index.md): Contextual explanations of claims, evidence, appraisal, boundaries and independence.

## Publications and WCC 1.0

- [Publication directory](https://witnessability.io/publications/index.md): Version-specific sources and the distinct status of Core, Model and individual Internet-Drafts.
- [Original Conceptual Core 1.0 Markdown](https://witnessability.io/downloads/WITNESSABILITY-CORE-1.0.md): Unchanged released text; SHA-256 382e39864e257d93a0afd16c377c54f59a2182ecc5c38178764ded29ffb3f034. The canonical reading page and full publication record are at https://witnessability.io/core/1.0/.
- [Witnessability Notation](https://witnessability.io/notation/index.md): The specification of the boundary-diagram notation (informative), version 0.4, as its original Markdown; the canonical page with the legend, four figures and the errata to version 0.4 is at https://witnessability.io/notation/ (errata at https://witnessability.io/notation/#errata), and the fixed page of version 0.4, for citation, at https://witnessability.io/notation/0.4/.
- [Witnessability Notation vocabulary](https://witnessability.io/ns/wn/index.md): The namespace document of the notation's RDF vocabulary (an informative bridge for notation 0.4), which states the namespace IRI; Turtle at https://witnessability.io/ns/wn.ttl and the JSON-LD context at https://witnessability.io/ns/wn.jsonld.
- [Release verification](https://witnessability.io/verify/index.md): Publication scopes, canonical ZIP, signed records and limits of an integrity finding.
- [Release key and trust boundary](https://witnessability.io/keys/index.md): Published fingerprint and the separate question of binding a key to an authority.
- [Errata and correction policy](https://witnessability.io/errata/index.md): Preserved errata records and the distinction between later clarification and immutable released bytes.

## Project context

- [About and authorship](https://witnessability.io/about/index.md): Research purpose, authorship, AI assistance, responsibility and focused contributions.
- [Project relationships](https://witnessability.io/ecosystem/index.md): The scopes of Witnessability, WEXP and CBEE.

## Structured records and verification procedure

- [Resource discovery manifest](https://witnessability.io/.well-known/ard.json): A static ARD catalog of the public release-verification skill, including its exact digest. DNS-AID points to this catalog; it does not provide a hosted agent, MCP server or A2A service. The procedure runs in the reader's own environment. ARD and DNS-AID are evolving draft specifications.
- [Legacy catalog format](https://witnessability.io/.well-known/ai-catalog.json): The same skill in the earlier AI catalog format for compatible readers.
- [Agent skill discovery index](https://witnessability.io/.well-known/agent-skills/index.json): Discover the public read-only release-verification skill and its content digest.
- [Signed release index](https://witnessability.io/downloads/RELEASE-1.0-INDEX.json): Version, status, release authority, all 20 original publication artifacts and component-license qualifications.
- [Release index signature](https://witnessability.io/downloads/RELEASE-1.0-INDEX.json.asc): Detached signature for the exact release-index bytes.
- [Publication checksum manifest](https://witnessability.io/downloads/SHA256SUMS-PUBLICATION.txt): SHA-256 values for the other 18 approved assets; the manifest and its signature are excluded from its own coverage.
- [Original external verification guide](https://witnessability.io/downloads/EXTERNAL-VERIFICATION-GUIDE.md): Published verification procedure, canonical ZIP baseline and internal 52-file manifest check.
- [Signed License Grant Record](https://witnessability.io/downloads/LICENSE-GRANT-RECORD-CORE-1.0.md): Component scopes, grant qualifications and the release's copyright limitations.