heypastelheypastel/pastel-outbound-system/skills/intent-radar/SKILL.md

intent-radar

Skill2 starsChanged 56 days agoMITView on GitHub

Radar — lists people Pastel caught doing something that suggests they could buy (asking for a tool, reacting to a competitor, discussing the problem, changing role), each with what they did and when. Use for "get me leads", "who's buying right now", "who engaged with this post", or building a warm list.

In this repository20 other files it gives its agents.
- AGENTS.mdAGENTS.md
- buying-committeeSkill
- call-readySkill
- campaign-launchSkill
- command-centerSkill
- competitor-watchSkill
- fit-checkSkill
- human-voiceSkill
- market-mapSkill
- message-writerSkill
- morning-briefSkill
- nudgeSkill
- outbound-briefSkill
- post-studioSkill
- priority-rankSkill
- reach-checkSkill
- reply-deskSkill
- social-warmupSkill
- what-worksSkill
- why-nowSkill

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Distinctive:
Every row answers one question: what did this person just do?
line 10
Distinctive:
Default list size: 30. The user can ask for more or fewer.
line 27
Distinctive:
No results → run the empty-result check from the house rules.
line 29

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

SKILL.md

Intent radar — Radar

House rules: ../command-center/shared/house-rules.md · Pastel tools: ../command-center/shared/pastel-tools.md

Every row answers one question: what did this person just do? The event and its date are the row.

Signal strength

Strength	What they did
🔥 Hot	Asked publicly for a tool or recommendation; complained about the problem we solve
🟠 Warm	Reacted to or commented on a competitor's or a category post; engaged with our own content; commented a keyword ("DEMO", "SALES", "interested") to claim someone else's giveaway
🟡 Early	New role in the last 90 days; team we'd equip is hiring; funding tied to this problem

Steps

Load the market map or outbound-brief.md.
Ask Pastel for the ranked list first: pastel_recommend_best_leads with the segment's filters (job_titles, countries, industries, icp_definition_id, and created_since when the user gives a time window). Use pastel_query_leads to go past the top of the list.
"Who engaged with X": find the post with pastel_query_posts, then pastel_leads_for_post.
Read what happened. Ranked results carry it in evidence (primary_event_type, primary_event_at, selection_reason). For the others, or to quote the actual comment, use pastel_get_details, five ids at a time.
Label each row 🔥 / 🟠 / 🟡 from the table. Anything older than 60 days drops a level. Read the comment itself: a one-word keyword under a "comment to get it" post shows interest in a free resource, not a buying decision, so it stays 🟠 even when the post is about exactly what we sell.
Default list size: 30. The user can ask for more or fewer.

No results → run the empty-result check from the house rules.

Output

# Radar — [what we looked for] — [date]
Source: Copilot ranking | lead search | post engagers · Rows: N · 🔥 n · 🟠 n · 🟡 n

| | Person | Headline | Company | What they did | When | lead_id |

Close with one pastel_show_leads call with the same ids in the same order, then hand the lead cards to fit-check.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free