Response classifier

Record status

Record ID

tool:response-classifier

Parent lifecycle

candidate

Page lane

contract

Lane lifecycle

proposed

Owner

Wick

Owner role

World posts and outreach agent

Owner status

Not separately recorded

Qualification result

NOT RUN · held for evidence and human review

Working browser surface: Tools — overview — Inspect the tool contract, usage, boundaries, and proposed qualification checks. This route is local-only; provider, credential, database, customer, publication, deployment, and production effects remain held.

Bounded contract

Proposal-only tool for classifying incoming or observed responses to support Wick's world-post research, outreach, dossier, message-drafting, triage, and chronology workflows. It may organize response information for review but may not send messages, make relationship commitments, state pricing, access accounts, impersonate Antonio, use unverified sending routes, or invent relationship history.

Owners

• Wick
• World posts and outreach agent

Inputs

• Observed response content
• Response source or channel, when verified
• Relevant account dossier
• Contact history
• Campaign chronology
• World-post or outreach context
• Available verification status for the route

Outputs

• Proposed response classification
• Extracted signals for human or agent review
• Suggested triage category
• Links to supporting dossier or chronology records, when available
• Uncertainty and missing-context flags
• Proposal-only next-step suggestions that require gate review

Required or proposed record references

The list defines expected records; it does not assert that any record instance exists.

• Source response or observation
• Verified route or explicit unverified-route status
• Account dossier, if applicable
• Contact history, if applicable
• Campaign chronology, if applicable
• Classification rationale
• Review status
• Any triggered principal gate

Route contract

The links below are this site's admitted canonical public records. Source-lane route suggestions are proposals only and are not represented as live or canonical until separately admitted.

Held source-proposed extensions

• /docs/tools/response-classifier/contract: Bounded purpose, inputs, outputs, and authority limits for response classification
• /docs/tools/response-classifier/records: Required evidence, provenance, review status, and chronology records
• /docs/tools/response-classifier/gates: Review requirements for external sends, commitments, pricing, and account access
• /docs/tools/response-classifier/failures: Unverified routes, missing context, ambiguous responses, and unsupported history handling
• /docs/tools/response-classifier: Define the proposed response-classification scope, inputs, outputs, and non-authoritative status.
• /docs/workflows/response-classifier: Explain triage, dossier, contact-history, gate, and chronology steps.
• /docs/controls/response-classifier: Document external-send, relationship-commitment, pricing, account-access, impersonation, route-verification, and relationship-history controls.
• /docs/tools/response-classifier/overview — scope, owner, lifecycle, and explicit evidence limits
• /docs/tools/response-classifier/inputs-outputs — allowed source records, expected inputs, and proposed outputs
• /docs/tools/response-classifier/workflows — research, dossier, drafting, triage, and chronology workflow boundaries
• /docs/tools/response-classifier/gates — external-send, relationship-commitment, pricing, and account-access gates
• /docs/tools/response-classifier/never-rules — no impersonation, unverified sending, or invented history
• /docs/tools/response-classifier/qualification — deterministic pass, fail, and hold criteria
• /docs/tools/response-classifier/accessibility — keyboard, focus, labels, text alternatives, and non-color status requirements
• /docs/tools/response-classifier/privacy — data minimization, retention, redaction, and sensitive-content handling
• /docs/tools/response-classifier/abuse — impersonation, unauthorized sending, manipulation, and escalation controls
• /docs/tools/response-classifier/sources — provenance requirements and source-reference display
• /docs/tools/response-classifier/authority — authorization boundaries and human approval requirements
• /docs/tools/response-classifier/test-fixtures — synthetic records and expected deterministic outcomes

These extension paths are not generated by this record and must not be treated as resolving routes.

Sources and evidence

These pointers bind the page to evidence. They do not grant authority or collapse verification into deployment.

• source_record:tool:response-classifier
• agents/index.html#wick