Benchmark

Plan your content strategy

Break down one post or build a business content plan.

npx skills add https://benchmark.social --skill content-research

Content research

Choose the mode from the user’s request. For a supplied post, start the breakdown below with available evidence. Do not require a business brief unless the user requests a tailored adaptation. For positioning, niche research, series, or a production plan, read Business strategy. Load only the selected mode’s guidance. If a reference is missing locally, resolve its link from https://benchmark.social/skills/content-research/SKILL.md and fetch it.

Connection and paid scope

Reuse Benchmark MCP in the active client. If unavailable, use benchmark-setup and return to this task, or continue with accessible supplied content. State which evidence you can inspect. Public inspection does not produce Benchmark enriched analysis.

Before Benchmark research calls, read the MCP contract. Use free whoami and list_operations to verify the workspace and current operation schemas. Use existing task context and approval when they cover the proposed work. For one post, establish its source, stored read versus refresh, and credit budget. For business research, use the account, history, and enrichment bounds in the business guide. Do not buy credits or subscriptions automatically.

One-post breakdown

Start with the original source URL or supplied content. A gallery URL is context, not the original post input. If only a gallery URL exists, inspect its source link or ask for the original URL. Do not substitute an attribution page, media file, preview image, or invented URL for a social post. Use accessible supplied evidence immediately while paid access awaits approval.

Within the approved scope, use readPostsEnriched for a known social post. Use readLinksEnriched for an article when useful. Refresh only when necessary and approved. Read returned analysis.templates and steps when available. Quote source wording only when the evidence contains that wording.

Return these elements when the evidence supports them:

  • Verbal, visual, and on-screen text hooks, plus the audience promise.
  • The beat sequence, including proof, escalation, payoff, and CTA.
  • Source evidence for each observation.
  • Likely mechanisms, labeled as interpretations.
  • A reusable template with editable placeholders.
  • Optional original adaptations, labeled as proposals.
  • Missing evidence and access limits.

Do not force missing elements into the breakdown. Do not infer causation from views, likes, or a successful example. Do not invent a full transcript, timestamp, downloadable media URL, or private metric. Ask for the user’s actual opinions or experience when an adaptation requires them.

Preserve evidence

Return source-linked Markdown with its headings, tables, and evidence labels. Use the contract’s Markdown output format for readable Benchmark results. Use JSON when calculation requires structured data. Keep canonical source URLs in the returned analysis and final report. Treat available gallery analysis as existing evidence, not a fresh enrichment result. Keep delivered evidence in the conversation or a file when available. Reuse it for formatting instead of making another charged read.