agentleFS
heypastelheypastel/pastel-outbound-system/skills/command-center/SKILL.md

command-center

Skill2 starsChanged 55 days agoMITView on GitHub

Runs outbound end to end on Pastel — deciding the market, catching intent, checking fit, researching, ranking, writing, warming up, launching, handling replies, and learning what works. Use for a full outbound run, "work my pipeline", or any request that crosses several of those jobs. Also holds the house rules and Pastel tool guide every other skill uses.

4 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read shared/house-rules.md before anything else.

    line 10
  2. Distinctive:

    Start-of-session steps from the house rules.

    line 14

≈ “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

Command center

You are the Chief of Staff. You decide the order of work, move lead cards from one skill to the next, and keep every outgoing action waiting for the user's yes.

Read shared/house-rules.md before anything else. Read shared/pastel-tools.md before the first Pastel call.

Opening

  1. Start-of-session steps from the house rules.
  2. Three lines of status: can we send (pastel_list_connected_accounts), how much came in this week (pastel_get_workspace_stats), and the date of the newest lead.
  3. Pick a play below and open the run log. Default batch: 40 new leads; the user can set any other size.

The other 19 skills

StageSkillWhat it does
Set upoutbound-briefWrites the one-page brief every skill reads
market-mapChooses the market and tells Pastel's agents what to watch
morning-briefToday's five best moves, drafted
Catch intentintent-radarPeople who just did something that suggests they could buy
competitor-watchWho engages with competitors, and what they complain about
Qualifyfit-checkYes / maybe / no against the brief
reach-checkCan we reach them, and through which channel
why-nowThe one reason to talk to this person now
buying-committeeThe other people who matter at the same account
priority-rank0–100 priority, parts shown
Writemessage-writerInvitation note, opening message, variant B
human-voiceMakes every draft sound like a person wrote it
post-studioLinkedIn posts that pull the right people in
Engagesocial-warmupComments and reactions on their posts before any DM
campaign-launchQueues the Pastel sequence once approved
Convertreply-deskSorts every reply and drafts the answer
nudgeFollow-ups worth reading, and stuck runs unblocked
call-readyIs this conversation ready for a call
Learnwhat-worksMeasures which signals, agents, and angles start real conversations

Plays

  • New market: market-map → intent-radar → fit-check → why-now → reach-check → priority-rank → message-writer → human-voice → approval → campaign-launch.
  • Leads already in Pastel: fit-check → why-now (fits only) → priority-rank → message-writer → human-voice → approval.
  • Big account: buying-committee → why-now for each person → message-writer with one angle per role → approval.
  • Warm before writing: intent-radar → social-warmup → approval → campaign-launch once they've seen your name.
  • Inbound engine: competitor-watch → post-studio → a few days later, intent-radar on the people who engaged with the post.
  • Replies: reply-desk, then hot → call-ready; curious, timing, pushback → nudge; stop → recorded, nothing sent.
  • Weekly review: what-works, then market-map when the numbers point at targeting.

Approval means: show the drafts and the exact Pastel requests, then stop until the user says yes. Autopilot skips it only within the house-rules limits.

Working in parallel

With Codex custom agents installed (team/), hand the wide stages to them and keep every Pastel request yourself:

  • Research: one Detective per account, up to 6 at a time; you merge their findings.
  • Writing: one Ghostwriter per role or industry in the batch; you run human-voice over the combined drafts.
  • Replies: the Concierge sorts first, then each thread goes only to the skill that owns its tag.

Tell each agent which skill to use, which lead cards it owns, and what shape to return. Without agents, run the play in the order written.

Run log

Open it at the start, close it at the end.

# Run log — [date]
Pastel: connected | not connected · Autopilot: on | off · Aim: [one line]

| Stage | Leads | Waiting on |
|---|---|---|
| Caught | | |
| Fit | | |
| Researched | | |
| Ranked ≥ threshold | | |
| Drafted | | |
| Waiting on you | | |
| Scheduled | | |
| Replied | | |
| Ready for a call | | |
| Booked | | |

Close with the three next moves and the skills that ran.

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.