agentleFS
guerrilla2799guerrilla2799/sdr-copilot/skills/handle-reply/SKILL.md

handle-reply

Skill1 starsChanged 4 months agoMITView on GitHub

Turn a prospect's reply into the next move — classify it, match objections to approved responses, and draft the in-thread answer in the rep's voice. Use when you say "they replied", "how do I respond to this", "handle this reply", or paste an email thread or LinkedIn response.

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 ${CLAUDE_PLUGIN_ROOT}/references/messaging.md and references/guardrails.md if not already in context.

    line 12
  2. Distinctive:

    Run the same check as write-email: would a busy buyer believe a specific human wrote this?

    line 54

≈ “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

Handle Reply

A reply is the most valuable event in outbound — don't waste it on a slow or generic answer. Input: the prospect's reply pasted in (ideally the whole thread for context, since there's no Outreach integration). If you paste only the reply, ask which sequence/hook it came from.

Before you draft

Read ${CLAUDE_PLUGIN_ROOT}/references/messaging.md and references/guardrails.md if not already in context. Look for voice-profile.md and learnings.md in the working folder — replies are where voice matters most (the prospect has now read two things from the rep; the third has to sound like the same person).

Step 1 — Classify the reply

One of seven types. Say which one and why before drafting:

  1. Interested — wants to talk, asks about a demo/trial
  2. Question — engaging but needs an answer first ("does it do X?", "how's it priced?")
  3. Objection — a reason not to engage ("we use the native tool", "we have the legacy incumbent", "no budget", "not a priority")
  4. Referral — "talk to [name] instead"
  5. Not now — timing deferral ("circle back in Q4", "mid-renewal")
  6. Negative / opt-out — "not interested", "remove me"
  7. Auto-reply / OOO — no human read it

Step 2 — Draft the response

Universal rules for reply drafts:

  • Shorter than the cold email. 1-3 sentences is the norm — the prospect already knows the pitch, don't re-pitch.
  • Same thread, no new links unless they asked for one.
  • Answer what they actually said first. A reply that ignores their words reads as automation and kills the thread.
  • One next step. The goal is a meeting (or a trial start for Champion types) — every draft moves toward one or gracefully exits.
  • Speed note: tell the rep replies should go out same-day, ideally within the hour. A warm reply cools measurably by tomorrow.

By type:

  1. Interested → Confirm and compress the path to the meeting: propose two concrete times or send the booking link, one sentence of what they'll get ("30 min, we'll pull up your actual environment if you want"). No victory lap.
  2. Question → Answer it straight in ≤2 sentences using only the messaging reference, then bridge to the meeting ("easier to show than describe — worth 15 minutes?"). Pricing specifics, contract terms, roadmap: do NOT answer (guardrail #7) — give the shape (the pricing model at a high level, that there's a real free tier) and route detail to the call.
  3. Objection → Match it to the approved response. Two sentences max, always end in a question. The map:
    • "The native/status-quo tool is fine" → the category-level move: it was built for an earlier era of the problem; ask where the newer/growing costs show up in it (they don't). Use the approved specifics in messaging.md.
    • "We have the legacy incumbent" → independence + renewal angle (Hook #1/#5 language). Ask when the renewal is — that's the real qualifying question.
    • "We built our own / DIY dashboards" → the Champion differentiator: "stop building and maintaining that yourself" — ask how much eng time it eats per month.
    • "No budget" → honest pricing + real free tier; the free tier IS the answer ("start free, buy it only when it's already working").
    • "Not my job" → ask who owns this, convert to a referral.
    • "We have a point solution" → point coverage vs. platform; do NOT overclaim against a point solution's core strength (banned claim) — position as "that solves one slice; who sees the whole picture?" Use the approved framing in messaging.md.
    • Anything not on this map → draft the best 2-sentence response from the differentiators, mark it [unapproved objection — flag to your marketing lead], and tell the rep to log it for the what-works review.
  4. Referral → Two drafts: (a) thank-you to the referrer asking for the intro ("mind looping them in? one line from you beats ten from me"), (b) the fresh first-touch to the new person referencing the referral. Then: "Say brief me on [new contact] before sending (b)."
  5. Not now → Accept it cleanly, anchor a specific comeback date in their words ("I'll come back the week of [date]"), offer the campaign's recordings/offer as the zero-pressure keep-warm (if one is live). Tell the rep to set the Outreach/HubSpot task for that date — a "not now" without a scheduled follow-up is a "never."
  6. Negative / opt-out → One gracious sentence, no counter-pitch, no "can I ask why." Tell the rep: mark opted-out in Outreach AND HubSpot today — this is a compliance action, not a courtesy.
  7. OOO → No draft. Note the return date and any delegate named; suggest a bump timed 1-2 days after return. If the OOO names a delegate who fits a persona, that's a lead — offer to brief them.

Step 3 — AI-writing check + log reminder

Run the same check as write-email: would a busy buyer believe a specific human wrote this? Replies are held to a higher bar than cold emails — the prospect is actually reading now.

Output format

**Reply type:** [1-7 name] · thread: [subject/channel] · persona: [Champion/Practitioner/Economic Buyer]

**Read:** [1-2 sentences — what they're actually saying, incl. tone]

**Draft:**
[the response]

**Why this move:** [one sentence]
**Log:** [what to record in HubSpot/Outreach — reply logged, task date, opt-out, etc.]

End with the next step: booking-link nudge, brief me on [referral name], or the follow-up date to set. Every handled reply is also data — remind the rep roughly weekly that what-works wants to know which objections keep coming up.

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.