Forecast your quarter from pipeline data
AI sales forecasting is an Agent Skill: a written method your AI applies on demand, in Zealos or in any client that reads SKILL.md.
AI sales forecasting is an Agent Skill: a written method your AI applies on demand, in Zealos or in any client that reads SKILL.md.
A 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.
| name | meddicc |
|---|---|
| description | The MEDDICC qualification framework and what each letter requires before a deal can advance. Use when assessing deal health, preparing a forecast, writing a deal review, or deciding whether a deal is real. |
| allowed-tools | none — this skill adds judgement, not reach |
| compatibility | Zealos, Claude Code, Cursor, Codex and any client that reads SKILL.md |
| license | Free to use and edit |
--- name: meddicc description: The MEDDICC qualification framework and what each letter requires before a deal can advance. Use when assessing deal health, preparing a forecast, writing a deal review, or deciding whether a deal is real. --- # MEDDICC A deal is qualified when every letter has evidence, not a guess. Say which letters are unevidenced rather than filling them in optimistically — a confidently-wrong qualification is worse than an obviously incomplete one. | | Requires | |---|---| | **M**etrics | A number the buyer stated, in their units. "Two days per rep per week", not "efficiency gains". | | **E**conomic buyer | A named person who can sign without asking anyone. If we have not spoken to them, say so. | | **D**ecision criteria | What they will compare us on, in their words and their order. | | **D**ecision process | The actual steps: security review, legal, procurement, board. With dates. | | **I**dentify pain | What breaks if they do nothing. If nothing breaks, there is no deal. | | **C**hampion | Someone who sells internally when we are not in the room, and who has something to gain. Tested by asking them to do something. | | **C**ompetition | Including "do nothing" and "build it ourselves", which win more often than any vendor. | ## How to report For each letter: **evidenced** (quote the source — call, email, note), **assumed** (say so), or **missing**. Then state the single highest-value unknown and the question that resolves it. One question, not a list. ## Red flags - Champion who will not introduce you to the economic buyer. - Metrics we authored rather than heard. - A decision process with no dates. - "Everyone loves it" with no named detractor. There is always a detractor.
The forecast conversation starts from which deals are evidenced rather than from whose gut everyone trusts most.
You can see the shape of the quarter without a weekly ritual, and the biggest single risk is named rather than felt.
The number arrives with its assumptions attached, which is what makes it usable for anything downstream.
It forecasts more consistently. It applies the same evidence bar to every deal, which is mostly what rep-to-rep variance is.
Stage history, close-date changes, activity recency and whatever qualification evidence exists. Close-date movement is the strongest single signal.
It replaces the part where everyone establishes what the numbers are, leaving the part where you decide what to do about them.
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-conventionsQualification 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.
/meddicc