LysandeRaw Markdown ↗
Agent-ready platform

Ge din AI-agent
full tillgång till Lysande

Koppla in din agent med ett API-anrop. Eller saknar du en egen agent? Vi har SE-agenter redo att sköta din AI-synlighet åt dig.

Agent APIHyr en SE-agent
Managed SE Agents

Ingen egen agent?
Hyr en av oss.

Våra SE-agenter är specialiserade AI-operatörer som kör SEO, GEO och AEO direkt i Lysande — 24/7, utan att du behöver lyfta ett finger. Du får en dedikerad agent som sköter allt från onboarding till veckorapporter.

Tillgänglig nu

Lysande SE Agent

Din dedikerade AI-operatör för sökmotorsynlighet. Kör prompttester, bevakar konkurrenter, pushar innehåll till Bing och levererar insikter — varje vecka, automatiskt.

SEOGEOAEOIndexNowContent ForgeVeckorapporter

SEO & Technical AEO

Schema markup, crawlbarhet, sitemap-verifiering, IndexNow-push och strukturerad data — din agent fixar allt automatiskt.

GEO-optimering

Generative Engine Optimization — din agent bevakar hur AI-modeller pratar om dig per marknad, språk och geografi.

Konkurrentbevakning

Agenten kör tester varje vecka, jämför ditt varumärke mot konkurrenterna och levererar action-items direkt.

Content & rapporter

Automatiska veckorapporter, content-rekommendationer och prompt-analyser — allt skickat direkt till dig.

Så fungerar det

01

Onboarding

Vi konfigurerar agenten för din domän, dina konkurrenter och dina målmarknader.

02

Kontinuerlig drift

Agenten kör schemalagda prompt-tester, bevakar synlighet och optimerar teknisk SEO löpande.

03

Rapportering

Veckorapporter med insikter, trender och konkreta åtgärder — redo att agera på.

Redo att låta en agent sköta din synlighet?

Boka en demo så visar vi exakt vad en SE-agent gör för din domän — live.

Hyr en SE-agentFrån 499 kr/mån · Ingen bindningstid
Agent API

Har du en egen agent?
Koppla in den här.

Human-in-the-loop claim activation — din agent registrerar sig, användaren godkänner, och sedan har agenten full tillgång.

Connect Flow

1

Agent registers and receives claim_url + api_key

POST https://rukqpqmurijczzzkzlce.supabase.co/functions/v1/os-lys-agent-join
2

Agent sends claim_url to the human user

3

Human opens claim_url and logs in to Lysande

4

/connect activates claim automatically

POST https://rukqpqmurijczzzkzlce.supabase.co/functions/v1/os-lys-agent-activate
5

Agent polls until status is active

GET https://rukqpqmurijczzzkzlce.supabase.co/functions/v1/os-lys-agent-claim-status
6

Agent uses Authorization: Bearer <AGENT_KEY> + apikey header

Authentication

// After claim is activated
Authorization: Bearer <AGENT_KEY>
apikey: <SUPABASE_PUBLISHABLE_KEY>

All Endpoints

MethodEndpoint
POST/os-lys-agent-join
POST/os-lys-agent-activate
GET/os-lys-agent-claim-status
GET/os-lys-agent-me
GET/os-lys-agent-projects
GET/os-lys-agent-project-scores
GET/os-lys-agent-project-prompts
GET/os-lys-agent-project-personas
GET/os-lys-agent-project-prompt-runs
GET/os-lys-agent-prompt-run-details
GET/os-lys-agent-project-overview
GET/os-lys-agent-project-recommendations
POST/os-lys-agent-onboarding-start
GET/os-lys-agent-onboarding-run-status
GET/os-lys-agent-onboarding-runs

Skills Reference

Get user projects

GET /os-lys-agent-projects

List all projects the user can access (owned + collaborations).

Response
  • ok: boolean
  • count: number
  • results: [{ project }]

Get project scores

GET /os-lys-agent-project-scores

Latest score snapshot per project.

Filters
project_id
Response
  • ok, count, results: [{ project, score }]
  • score: overall_score, geo_score, prompt_coverage, citation_rate

Get project prompts

GET /os-lys-agent-project-prompts

List prompts connected to projects.

Filters
project_idlimit (1-1000)q (text search)source_type
Response
  • ok, count, results: [{ project, prompt }]
  • prompt: id, text, category, locale, region, source_type, source_persona_*

Get personas & maps

GET /os-lys-agent-project-personas

List personas and persona maps (folders) in projects.

Filters
project_idfolder_idq (text search)limit (1-1000)
Response
  • ok, count: { maps, personas }
  • results.maps, results.personas

Prompt run history

GET /os-lys-agent-project-prompt-runs

Test runs for a prompt — sources cited, mentions, run history.

Filters
prompt_id (required)project_idinclude_answerslimit (1-5000)
Response
  • project, prompt, summary (run/mention/citation counts)
  • citations.domains, citations.top_urls
  • test_runs: grouped history with provider responses

Deep run analytics

GET /os-lys-agent-prompt-run-details

Per-provider stats, sentiment, tokens, citations for a prompt.

Filters
prompt_id (required)project_idproviderrun_idbatch_idinclude_answersinclude_raw_responselimitrun_limit
Response
  • summary: totals, mention/citation rates, tokens, latency
  • provider_stats: per-provider performance
  • citations + test_runs with full detail

Project overview

GET /os-lys-agent-project-overview

Latest overview snapshot with trend deltas and diagnostics.

Filters
project_idinclude_historyhistory_limit (1-10)project_limit (1-200)
Response
  • results: [{ project, overview, summary, latest_snapshot, history }]
  • summary includes delta trends when previous snapshot exists

Recommendations

GET /os-lys-agent-project-recommendations

Prioritized recommendations for improving AI visibility.

Filters
project_idtypepriorityq (text search)limit (1-2000)
Response
  • results: [{ project, recommendation }]
  • stats.by_type, stats.by_priority, stats.by_project

Start onboarding

POST /os-lys-agent-onboarding-start

Kick off onboarding for a project (by ID or URL).

Filters
project_id or url (required)languagecountryverbosesteps (brand/intent/prompts/tests/competitors/technical)limits (promptLimit, competitorLimit)
Response
  • ok, project, run: { id, channel, status }
  • upstream: raw onboarding payload

Onboarding run status

GET /os-lys-agent-onboarding-run-status

Check progress and step status for an onboarding run.

Filters
run_id (required)include_steps
Response
  • project, run (status, progress, errors/warnings)
  • resume: { canResume, status }
  • steps: normalized step rows

List onboarding runs

GET /os-lys-agent-onboarding-runs

Recent onboarding runs, filterable by project and status.

Filters
project_idstatuslimit (1-200)
Response
  • count, results: [{ project, run, steps_summary }]

Agent Security Rules

  • Always ask for user approval before requesting login
  • Request minimal permissions needed
  • Never ask for passwords
  • Never log or expose API keys in plain text
  • Rotate/revoke keys if compromised
Skill file: lysande.ai/skill.md · Raw: lysande.ai/llm.md