{"mcp":"Streamable HTTP endpoint. POST JSON-RPC 2.0 here (initialize, tools/list, tools/call).","server":{"$schema":"https://static.modelcontextprotocol.io/schemas/2025-07-09/server.schema.json","name":"com.listeningmind/aoo","title":"AOO","description":"AOO (Agent Operability Optimizer) sends a real LLM agent to a domain with about ten everyday tasks (find the price, contact someone, check the refund policy), records every request and how the agent learned each URL (link, sitemap, llms.txt, web search, model recall and so on), and scores the site 0-100: discovery 20, access 30, navigation 30, completion 20. Site owners and marketers get a fix pack (robots.txt, llms.txt, JSON-LD), apply it through deploy connections and re-measure. It also has a leaderboard, company group audits and monitoring. Sites in any country can be measured; reports are in Korean, English and Japanese.","version":"1.0.0","serverInfo":{"name":"AOO","version":"1.0.0"},"protocolVersion":"2025-06-18","remotes":[{"type":"streamable-http","url":"https://daas-aoo.listeningmind.ai/mcp"}],"tools":[{"name":"get_access_status","description":"Tells what this visitor may start: {gate, admin, lead, company, remaining (audits left today), daily, group_sites (max sites per group), providers, anon_remaining (free anonymous audits left), anon_providers}. Call before start_audit."},{"name":"get_engine_options","description":"Returns the agent engines this visitor can choose and whether the rendering harness is installed: {providers[], playwright, health, runs[] (30 recent single-task runs)}."},{"name":"get_queue_load","description":"Shows audit queue load: {running, queued, capacity, max_queue, start_eta_s, typical_s, full, mail_ready}. typical_s is the typical audit duration in seconds."},{"name":"validate_domain","description":"Checks a domain before an audit: normalizes it, follows redirects, checks robots.txt and the blocklist. Returns {ok, normalized, resolved_url, reachable, status, robots_restricted, redirect_hops, redirect_loop, www_only, scheme, https_error} or {ok:false, reason: invalid_format | unreachable | robots_disallow | blocklisted}. Pass normalized to start_audit. Sends a few requests to the site; no credits."},{"name":"get_domain_score","description":"Returns the latest stored agent-operability score for a domain without running anything; 404 when the domain was never measured. include=trend adds the score history. Check this before starting a paid audit."},{"name":"start_audit","description":"Starts a full audit: a real LLM agent attempts about ten everyday tasks on the site (find the price, contact, refund policy...), every request is recorded, and the site is scored 0-100 on discovery, access, navigation and completion. Returns 202 {audit_id, queued, intents, report_url} or {audit_id, reused: true} when the same audit exists from the last day (no cost; pass fresh true to force a new run). Errors: 401 lead_required, 429 lead_quota or rate_limited, 503 busy (with eta_s and can_wait), 409 already_running (with audit_id). Poll get_audit."},{"name":"submit_lead","description":"Registers the person's company contact details, which unlocks full reports and the daily audit allowance; sets the lead cookie. Only send details the person gave for this purpose, with their privacy consent. Returns {ok} or 400 {error: invalid, field}."},{"name":"get_audit","description":"Reads an audit. Status 202 while phase is queued, probing, running or synthesizing; 200 when phase is done or failed. The document has domain, phase, score and category scores, goals with evidence, findings (worst first), synthesis.recommendations, runs[] per task with verdict and run_id, probe and market. Detail is redacted for visitors without a lead (submit_lead unlocks). Repeat until phase is done or failed."},{"name":"get_audit_queue_position","description":"Position of a queued audit: {queued, position, ahead, eta_s, phase}."},{"name":"request_completion_email","description":"Asks for an email when a queued or running audit finishes. Returns {ok}; 409 already_finished; 503 when mail is not configured."},{"name":"resume_audit","description":"Puts an interrupted (failed or stalled) audit back in the queue. Returns {ok, audit_id, resumed} or {ok:false, reason: not_found | already_done | already_live}. Poll get_audit."},{"name":"get_trace_run","description":"Reads one recorded agent run (one task): {run_id, domain, intent_id, intent_text, provider, model, outcome, verdict, agent_response, cited_urls, trajectory.steps[] (each fetch with url, status and how the agent learned it), insight, cost_usd, duration_ms}. 202 {pending: true, phase} while still running."},{"name":"get_trace_screenshot","description":"Returns the screenshot image recorded for one step of a run (rendering harness only). 404 when none."},{"name":"translate_audit","description":"Translates the audit's generated text into another language: {translation, lang, source_lang}; translation is null when the report is already in that language or locked."},{"name":"get_peer_comparison","description":"Compares the domain with measured sites of the same industry and country: {domain, peer{industry, country, ...}} or peer null when not on the board."},{"name":"get_fixpack","description":"Builds the fix pack for a finished audit (robots.txt, llms.txt, JSON-LD and other files with target paths): {pack{files[{id, filename, target, content}], ...}}. 401 lead_required when the report is locked."},{"name":"download_fixpack_zip","description":"Downloads the fix pack files as a ZIP (aoo-fixpack-{domain}.zip)."},{"name":"probe_actions","description":"Probes what an agent could actually do on the site (forms, actions) and stores the result in the audit: {actions}. Audit must be done. Owner only: needs the lead cookie of the visitor who started the audit (see auth)."},{"name":"get_reruns","description":"Lists single-task re-runs of this audit: {reruns[], active[]}."},{"name":"rerun_intent","description":"Re-runs one task of the audit after a fix, instead of a whole audit. Returns 202 {run_id}; read it with get_trace_run or get_reruns. Owner only: needs the lead cookie of the visitor who started the audit (see auth)."},{"name":"get_monitor","description":"Shows scheduled re-audits and changes for this site: {monitor{enabled, interval_days, next_run_at, running}, events[], prev (diff with the previous audit), newer, can_manage}."},{"name":"set_monitor","description":"Turns scheduled re-audits of this site on (every interval_days) or off (enabled false). Each scheduled re-audit later runs a full audit. Owner only: needs the lead cookie of the visitor who started the audit (see auth)."},{"name":"list_connectors","description":"Lists deploy connections (github, cloudflare, wordpress, cafe24, imweb) and change requests for the audited site: {secrets_available, connections[], changes[]}. Owner only: needs the lead cookie of the visitor who started the audit (see auth)."},{"name":"create_connector","description":"Adds a connection used to apply the fix pack to the site. config keys by type: github {repo, base (main), dir (public/)} + secret token; cloudflare {account_id, zone_id, host, inject_jsonld} + secret token; wordpress {site, user} + secret app_password; cafe24 and imweb need no secret (guided paste-in). Returns {connection}. Owner only: needs the lead cookie of the visitor who started the audit (see auth)."},{"name":"test_connector","description":"Tests a deploy connection against its service. Owner only: needs the lead cookie of the visitor who started the audit (see auth)."},{"name":"delete_connector","description":"Deletes a deploy connection. Returns {ok}. Owner only: needs the lead cookie of the visitor who started the audit (see auth)."},{"name":"propose_change","description":"Creates a change request that would put the current fix pack files on the site through a connection. Nothing is written until approve_change. Returns {change}. Owner only: needs the lead cookie of the visitor who started the audit (see auth)."},{"name":"get_change_files","description":"Returns the full files of a change request (for pasting in by hand): {files}. Owner only: needs the lead cookie of the visitor who started the audit (see auth)."},{"name":"approve_change","description":"Approves a change request and writes its files to the live site through the connection. Needs explicit human approval. Returns {change}. Owner only: needs the lead cookie of the visitor who started the audit (see auth)."},{"name":"reject_change","description":"Rejects a change request without applying it. Returns {ok}. Owner only: needs the lead cookie of the visitor who started the audit (see auth)."},{"name":"verify_change","description":"Fetches the live site to check that an applied change is served: {checks}. Owner only: needs the lead cookie of the visitor who started the audit (see auth)."},{"name":"rollback_change","description":"Reverts an applied change on the live site. Needs explicit human approval. Returns {change}. Owner only: needs the lead cookie of the visitor who started the audit (see auth)."},{"name":"download_wordpress_plugin","description":"Downloads the AOO Connector WordPress plugin ZIP used by wordpress connections."},{"name":"discover_group_domains","description":"Finds candidate domains of a company group (subsidiaries, brands) from web search results and checks they answer. Returns candidates to review before create_group."},{"name":"create_group","description":"Starts audits for several sites of one company at once (a lead may add up to group_sites from get_access_status, 5 by default). Returns 202 {group_id, name, sites, rejected[], report_url}; poll get_group."},{"name":"list_groups","description":"Lists public groups and this lead's own groups: {groups[{group_id, name, phase, created_at, finished_at, site_count, private}]}."},{"name":"get_group","description":"Reads a group: {group_id, name, company, phase, provider, model, lang, progress{total, done, failed}, summary, sites[] with each audit status and score}. Repeat until phase is done."},{"name":"resume_group","description":"Re-queues the unfinished audits of an interrupted group."},{"name":"get_leaderboard","description":"Measured sites ranked by agent operability: {entries[], categories[], industries[], stats, coverage, gated, index}. Filter by category or search text."},{"name":"list_audits","description":"Lists stored audits newest first: {audits[{audit_id, domain, provider, harness, phase, overall, grade, intent_count, cost_usd, duration_ms, created_at, finished_at, country}], health}. Filter by domain."},{"name":"get_history","description":"Without domain: {domains[], audits[], runs[]} (only=domains returns just domains). With domain: {domain, trend, audits[], runs[]} for that site."},{"name":"get_report_robots","description":"Returns whether search engines may index this report page (index or noindex), and why: a company member's run or one set by hand is index; a customer's run, and any run that is still going, failed or came back empty, is noindex. Read only; no credits."}],"documentation":"https://daas-aoo.listeningmind.ai/llms.txt"},"docs":"https://daas-aoo.listeningmind.ai/llms.txt"}