Insaight — Post Research

Turn a single LinkedIn post into a map of the people around it. The post is usually the least interesting part: the comment thread is where prospects self-identify — describing their own pain in their own words, naming the tools they already pay for, arguing with the author. Read the thread as a lead list, not as a conversation.

Shared conventions (all Insaight skills):

• Call insaight:get_config() once per session for the user's Notion pages and company name/slug. If it reports unconfigured: true, ask the user to edit the file at the returned path before saving to Notion.
• Read cheaply: list_accounts → list_posts (slim index) → get_posts on the few URNs worth reading (max 20 per call). list_people / list_comments are free; the scrape_* tools call Apify and cost money — only scrape when data is missing or stale.
• Engagement benchmarks (adjust for the niche): < 10 likes low, 10–40 normal, 40+ high signal — read those in full.

Workflow

Step 1 — Get the post URL

Either form works:

• https://www.linkedin.com/posts/AUTHOR_SLUG-activity-ID-TOKEN
• https://www.linkedin.com/feed/update/urn:li:activity:ID

If the user only described the post, ask for the URL — comment scraping is keyed on it, and there is no reliable way to search your way back to a specific post from a description.

Step 2 — Load the post itself

Check what is already stored before spending anything:

insaight:list_posts(account="author-slug")

If the post isn't tracked, scrape the author. This stores their whole recent feed, so any later work on the same author is free:

insaight:scrape_profile(url="author profile or company page URL", max_posts=30)

Then read the full text:

insaight:get_posts(urns=["urn:li:activity:..."])

Read the post before the comments. The same remark ("we've been fighting this for months") means something different under a hiring announcement than under a product launch, and you can only grade the comments against what prompted them.

Step 3 — Scrape the comments

insaight:scrape_post_comments(post_url="...", max_items=100, profile_mode="short")

profile_mode="short" is free and already carries each commenter's name and headline — usually enough to judge seniority. Step up to "main" ($0.002/profile) only when the headlines leave you unable to tell who these people are.

Leave include_replies=True (the default). Replies hold the candid takes: people hedge in a top-level comment and say what they actually think two levels down.

Step 4 — Read the comments

insaight:list_comments(post_url="...", limit=100)

This reads from the local DB and costs nothing, so re-query freely. On a busy thread (50+ comments), call once with min_likes=2 to see what the audience itself endorsed, then again unfiltered — a quiet, unliked comment from a senior buyer is often the most valuable line in the thread, and a likes filter alone would hide it.

Step 5 — Enrich selectively

When a commenter looks like a genuine prospect but their headline leaves the deciding question open (do they own a budget? is their company the right shape?), pull the full profile:

insaight:scrape_person_profile(url="commenter LinkedIn URL")

Pick the 1–3 that matter. Blanket-enriching a hundred-comment thread burns Apify credits on data nobody will ever act on.

Step 6 — Write the brief

Use these headings:

📌 Post Summary

• Author, role, company
• Date posted, language
• Engagement: likes / comments / shares (see shared benchmarks above)
• One-line gist of what the post is about

💬 Comment Themes

Group the comments into 3–6 themes — e.g. agreement on a pain point, pushback on the approach, vendor recommendations, personal anecdotes. For each: the theme, a representative quote with attribution, and how many comments fall under it. Theme sizes tell the user which reaction is the market's, not one loud person's.

🌟 Notable Commenters

The top 5–10 by comment engagement or seniority. For each: name, headline, why they stand out, their comment in a line, and the next step you'd take (research, enrich, reach out, ignore).

🎯 Buying / Pain Signals

Quote the comments that reveal active evaluation ("we're looking at…", "thinking about…"), frustration with the status quo, tools they currently use, or budget and timing hints. Quote them rather than summarizing — the user's next message will borrow their words.

🤝 Warm Leads

Commenters worth contacting, ranked by fit against the post's topic and their engagement. Give each one sentence naming the hook this thread hands you.

⚔️ Competitor / Vendor Mentions

Every tool, product, or company named in the comments — alternatives and complements both.

❓ Open Questions

What you'd want to know but couldn't determine from the thread alone.

Step 7 — Save to Notion

Same pattern as the other research skills: save under [NOTION_RESEARCH_PAGE] with the title Post — [author last name] — [topic gist] — [YYYY-MM-DD]. Check for an existing page on the same post first and update it rather than creating a duplicate.

Judgment calls

• Thin threads: under ~5 comments, the discussion has little to say. Center the brief on the post and note who engaged, and tell the user plainly that there is no lead list here instead of padding one out.
• Author replies: whoever the author bothers to answer is someone they value — a possible warm-intro path, and a read on the author themselves.
• Sentiment shifts: a thread that turns from agreement to pushback partway through marks a real fault line in how the market sees the topic. Call it out.
• Names beat adjectives: "three commenters named [tool] as what they use today" is something the user can act on; "some interest in tooling" is not.