Kept AI
Back

Content Creation

Choose evidence-backed content channels

Use recorded citation patterns to plan channel-specific content experiments.

Connect your Kept MCP
Kept AIAI assistantYour workspace
Open Kept AI

Command

Use list_projects({}) to confirm my supplied project host is visible. Use my chosen days value of 7, 30, or 90 for every time-windowed call. Read list_sources({host, kind: "domains", days, limit: 100}) and list_sources({host, kind: "urls", days, limit: 100}). Inspect get_gap_analysis({host, days}) and selected get_prompt_runs({host, prompt_id, days}) after list_prompts({host, limit: 100}). Group observed sources into channels, marking uncertain classifications. Recommend three channel experiments with audience, angle, and success criteria. Draft one example using my approved facts. Scheduling or publishing requires an independent connection and review; citations alone do not prove channel effectiveness. Return evidence links, coverage limits, and explicit unknowns; never infer unsupported scores.

What you need

  • A visible host, supported window, audience, and approved messaging.
  • Provide source text for content inspection; publishing and scheduling require independent connections.
Read the setup guide

Map observed citations to channels

Compare cited domains and URLs to understand which channels appear in sampled answers. Classify sources transparently and retain uncertain entries, since an appearance in the citation record does not establish editorial quality, audience reach, or a causal advantage for that channel.

Turn patterns into testable briefs

Translate the source map into a small set of content experiments with distinct audiences and formats. Each brief includes a proposed angle, supporting evidence, review requirements, and a measurement plan that your team can execute through its own publishing tools.