Post-call automation turns a recorded conversation into the three things that outlive it: the commitments each side made, the fields that changed, and the follow-up that has to go out today.
/call-recap-formatA pre-call brief is worth writing only if it changes what you ask: who is in the room, what they are measured on, what happened last time, and the one question you cannot afford to leave without answering.
/discovery-questionsAn AI SDR agent is only as good as the definition of a good customer it is working from, which is why the fit criteria belong in a written skill rather than buried in a prompt nobody can find.
/our-icpAI email triage is worth having only when it sorts by consequence rather than by recency: what blocks a deal, what someone is waiting on you for, and what can be read at the end of the week.
/email-rulesCRM data hygiene survives AI only if one rule is written down first: a field left blank is a known unknown, while a field filled in by inference is an unknown unknown that will be reported on as fact.
/crm-field-conventionsAccount research earns its time only when it changes the first sentence you write, which means looking for what recently changed rather than assembling a profile anyone could have found.
/our-icpQualification is a claim about evidence, not a score: a deal is qualified when every criterion has a source you can quote, and unevidenced criteria are named rather than filled in optimistically.
/meddiccA follow-up gets a reply when it opens with the specific commitment made on the call, stays under 120 words, and asks for exactly one thing the recipient can answer in a sentence.
/email-rulesA writing-voice skill is the highest-leverage thing you can give an AI, because it changes every sentence the product produces and it needs no tools, no integrations and no data.
/our-writing-voiceAI writing gives itself away structurally before it does lexically: paragraphs of identical length, lists that always contain three items, and a closing summary that restates a body the reader just finished.
/no-slopA pipeline review is useful in proportion to how uncomfortable it is: the job is to find the deals that are slipping quietly, not to narrate the ones everybody already knows about.
/meddiccA pricing objection is almost never about price: it is a value problem, a comparison to something cheaper that does less, or an internal conversation the buyer has not told you about yet.
/objection-handlingCompetitive positioning works when you concede what is true first: naming what a competitor does well buys credibility for everything after it, and the buyer already knows anyway.
/competitor-positioningA proposal is read by people who were on none of the calls, so it has to open with what the buyer told you, in their words — get that visibly right and it buys the read for everything after it.
/proposal-structureAutomatic call scoring is only fair if the rubric is visible: score the observable things — whether a next step was agreed, whether questions were about the past — and leave judgement to a human.
/discovery-questionsAn account plan is worth the hour only if it names the one unknown that would most change your approach — everything else in a plan template is decoration you will never re-read.
/our-icpAfter-call work is the twenty minutes nobody books: the CRM update, the recap, the follow-up, the task for the thing you promised — and it is the first work to disappear on a busy day.
/call-recap-formatA cold sequence works when each touch carries new information: the second email should not exist unless it says something the first one did not, and most sequences fail this test by touch three.
/email-rulesA connection note has about three hundred characters and one job: prove you know who they are, without the sentence that reveals you found them in a filtered list.
/our-writing-voiceA Gmail skill tells the model how to search an inbox the way a person would — by thread and by relationship rather than by keyword — which is the difference between a search tool and an answer.
/email-rulesEnrichment is worth automating only with a rule about what not to write: a contact record improved with three verified fields is more useful than one filled completely with two guesses.
/crm-field-conventionsPulling opportunity context is a summarisation problem with a trap: the useful answer is what changed and what is missing, not a recitation of fields the reader can already see.
/meddiccAn automated standup is only read if it is short and specific: what moved, what is blocked, and what nobody has touched — with the last category being the one that justifies posting at all.
/call-recap-formatA useful Zoom summary is not a compression of the transcript: it is the commitments, the decisions and the unresolved risks, written for someone who was not on the call.
/call-recap-formatThe hard part of posting consistently is not writing, it is noticing: the week already contained something worth saying, and a drafting skill mostly needs to know what counts.
/our-writing-voiceA QBR earns the meeting only if it says something the customer does not already know, which means leading with what changed on their side rather than with your usage charts.
/proposal-structurePre-meeting research has a five-minute budget, so it should return exactly three things: what changed recently, who you are talking to, and the one thing that would embarrass you not to know.
/our-icpDeduplication goes wrong when merging is treated as the goal: the safe operation is to propose a merge with the evidence for it, because an incorrect merge is materially harder to undo than a duplicate.
/crm-field-conventionsA forecast is a claim about evidence, so the useful output is not a number but a split: what is evidenced, what rests on an assumption, and which single deal moves the quarter most if it slips.
/meddiccOnboarding fails on dependencies rather than on effort, so the skill worth writing is the one that names what you need from the customer, by when, and what happens to the date if it does not arrive.
/call-recap-formatAnswering from documentation is safe only with an explicit rule about the boundary: say what the docs say, cite where, and escalate rather than reconstruct an answer that is not written down anywhere.
/our-writing-voiceGuardrails matter more than limits: the rule that protects a deal is not the maximum discount but the requirement that no number, term or date is ever committed without a human.
/pricing-guardrailsGood discovery asks about what happened rather than what the buyer wants, because people are reliable historians of their own problems and unreliable architects of their own solutions.
/discovery-questionsA stalled deal is rarely revived by a check-in: it needs a reason for the buyer to spend attention now, which means new information rather than a reminder that you are still here.
/objection-handlingA folder containing a SKILL.md file: metadata that says what the skill does and when to use it, plus instructions the AI loads only when a task matches. It is an open format, so the same skill works in Zealos, Claude Code, Cursor and Codex.
An agent is the job; a skill is the method. Rewrite this call into a blog post is an agent. Our writing voice and MEDDICC are skills the agent applies while doing it. Most skills carry no tools at all.
Not to start. Every workspace gets twelve built-in skills covering writing voice, qualification, CRM conventions and objection handling, and each is a starting point you edit rather than a rule you inherit.
No, and this is the property that makes installing one safe. A skill can narrow which tools an agent may use but can never grant one, so a skill applied to a read-only agent stays read-only.