agentleFS
marichalnickmarichalnick/sales-skills-for-claude/skills/sales-router/SKILL.md

sales-router

Skill0 starsChanged 28 days agoMITView on GitHub

Entry point for sales work. Use when the user asks what to do next with a prospect or account, or wants to prepare, follow up, qualify, handle an objection or log a sales call without naming a specific skill. Routes a request to the right skill (signal-verified-outreach, discovery-call-prep, objection-handler, call-to-crm) and chains them across a deal, from first touch to CRM update.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 10 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    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. Distinctive:

    If the request fits two, do them in deal order: outreach → prep → objection → CRM.

    line 19
  2. Distinctive:

    If it fits none, help directly and say which skill would be closest.

    line 20
  3. Distinctive:

    Look at what exists for that account and answer with one concrete next action:

    line 37
  4. Distinctive:

    No verified trigger yet → research and verify one, or recommend waiting.

    line 38
  5. Distinctive:

    Call done, notes available → CRM update and follow-up.

    line 41
  6. Very specific:

    Look for ACCOUNT-<name>.md and PREP-<name>.md in the working folder.

    line 45
  7. Distinctive:

    If two files disagree, do not silently pick one: show the conflict and ask.

    line 48
  8. Distinctive:

    Keep unknowns as unknowns.

    line 49
  9. Distinctive:

    Write new facts back only after verifying them or after the user confirms them.

    line 50
  10. Distinctive:

    No file found: start fresh and say so.

    line 51
  11. Distinctive:

    Verify before you claim. Unknown is an acceptable answer.

    line 54

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Sales Router

Pick the right workflow, and carry context from one step to the next.

Routing

The user wants…Use
To contact a company or person, a cold email, a batch of outreach, a follow-upsignal-verified-outreach
To prepare a call or meetingdiscovery-call-prep
To answer pushback, live or writtenobjection-handler
To process a call: transcript, notes, recap, CRM, handoffcall-to-crm

If the request fits two, do them in deal order: outreach → prep → objection → CRM. If it fits none, help directly and say which skill would be closest.

The deal flow

signal-verified-outreach  →  ACCOUNT-<name>.md   (verified triggers, stakeholder)
        ↓
discovery-call-prep       →  PREP-<name>.md      (hypotheses, questions, MEDDICC gaps)
        ↓
     [the call]  ←  objection-handler when pushback comes up
        ↓
call-to-crm               →  CRM update, handoff, follow-up email

Each step reads the previous file if it exists, so facts are verified once and reused, not re-researched or re-invented.

"What should I do next with [account]?"

Look at what exists for that account and answer with one concrete next action:

  • No verified trigger yet → research and verify one, or recommend waiting.
  • Outreach sent, no reply → a follow-up with something new, or a different stakeholder.
  • Meeting booked → prep brief.
  • Call done, notes available → CRM update and follow-up.
  • Gaps in MEDDICC → which question to ask, and to whom.

Context resolution protocol

  1. Look for ACCOUNT-<name>.md and PREP-<name>.md in the working folder.
  2. Load only facts marked Fact with a source. Treat Inference as a hypothesis, never as a fact.
  3. Re-check the age of every trigger. Older than ~90 days: context, not a reason to reach out.
  4. If two files disagree, do not silently pick one: show the conflict and ask.
  5. Keep unknowns as unknowns. Never fill a gap to make a file look complete.
  6. Write new facts back only after verifying them or after the user confirms them.
  7. No file found: start fresh and say so.

Shared rules

  1. Verify before you claim. Unknown is an acceptable answer.
  2. Hypotheses are questions, not statements about the prospect.
  3. A human approves anything that leaves the building.

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.