agentsclimarketplace

Meeting outcome logger

Skill stephenrogan/csm-skills/skills/meeting-outcome-logger

Structures what happened in a customer meeting into a CRM-ready activity record with outcome classification, sentiment assessment, next steps, and follow-up triggers. Bridges the gap between "the meeting happened" and "the system knows what happened." Use when asked to log a meeting outcome, create a CRM activity record, document what happened in a call, classify a meeting result, or when any customer interaction needs to be recorded in a structured format for downstream use. Also triggers for questions about meeting documentation, activity logging, CRM record creation, or structured meeting capture.From its SKILL.md

Install
npx -y skills add stephenrogan/csm-skills --skill meeting-outcome-logger

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its file declares

Copied from the file, not written here

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

8.2 KB, ~1.5k tokens by cl100k_base, as published. Nobody here has run it

Meeting Outcome Logger

Structures a customer meeting into a CRM-ready record. Different from the call summary generator (which extracts discussion points and action items from notes) -- this skill classifies the meeting outcome, assesses sentiment, and produces a structured record optimised for CRM logging and downstream reporting.

The gap this fills: most CSMs either skip CRM logging entirely (too time-consuming) or write a sentence that does not capture enough for anyone else to act on. This skill makes logging take 2 minutes instead of 10 and produces a record that is genuinely useful.

How to Use

After a customer meeting, provide:

  • Account name and meeting type (check-in, QBR, escalation, onboarding, renewal, ad hoc)
  • Attendees (customer and internal)
  • Duration
  • What happened (brief description -- a few sentences or bullet points)
  • What was decided or agreed
  • What actions were committed (by you and by them)
  • How the customer seemed (positive, neutral, frustrated, disengaged, enthusiastic)
  • Anything that changed about the account status (new risk, new opportunity, new stakeholder, resolved issue)

Outcome Classification

Every meeting produces one of these outcomes. The classification drives downstream workflows:

OutcomeDefinitionDownstream Trigger
Positive progressMeeting advanced the relationship or a specific initiative. Customer engaged, actions agreedLog activity. Update health trajectory if warranted
Status quoRoutine touchpoint. No significant change in either directionLog activity. Standard cadence continues
Concern surfacedCustomer raised an issue, frustration, or risk signal during the meetingLog activity. Flag for risk assessment. CSM to investigate
Escalation neededIssue identified that requires internal team involvementLog activity. Trigger escalation workflow
Expansion signalCustomer expressed interest in growth (more users, higher tier, new use case)Log activity. Flag expansion opportunity
Relationship repairMeeting addressed a prior issue. Trust rebuilding in progressLog activity. Update risk status if improving
Customer disengagedCustomer was present but minimally engaged. Short answers, low energy, delegationLog activity. Flag as engagement risk. CSM to assess
Commitment madeSpecific commercial or operational commitment from either sideLog activity. Create commitment record for tracking
No-showCustomer did not attend the scheduled meetingLog no-show. Flag for CSM follow-up. If repeat pattern, flag as engagement risk

Sentiment Capture

Beyond the outcome classification, capture the emotional tone:

SentimentIndicatorsWhat It Means for the Account
EnthusiasticVolunteered positive feedback, discussed expansion, asked about new features, forward-looking languageStrong health signal. Potential advocacy candidate. Look for expansion timing
SatisfiedProfessional, engaged, answered questions fully, no complaintsHealthy baseline. Standard management
NeutralTransactional, brief answers, no strong signal in either directionNot a problem yet, but not building momentum either. Monitor for drift
ConcernedRaised specific issues, asked pointed questions about timelines or commitments, referenced alternativesEarly risk signal. Address the concerns before they become decisions
FrustratedExpressed dissatisfaction directly, referenced prior failures, impatient with answersActive risk signal. Escalate if not already addressed. Do not wait for the next meeting
DisengagedShort meeting, minimal contribution, delegated to a junior person, cancelled or shortened the meetingRelationship cooling signal. This is often a leading indicator of churn -- the customer stops caring before they stop paying

CRM Record Format

## Meeting Record: [Account Name]
**Date:** [date] | **Type:** [meeting type] | **Duration:** [minutes]

**Attendees:**
- Customer: [names, roles]
- Internal: [names, roles]

**Outcome:** [classification from above]
**Sentiment:** [assessment from above]

**Summary:** [2-4 sentences capturing what was discussed and concluded]

**Decisions Made:**
- [Decision 1]
- [Decision 2]

**Action Items:**
| Action | Owner | Deadline |
|--------|-------|----------|
| [action] | [owner] | [date] |

**Account Status Change:** [None / New risk: [detail] / New opportunity: [detail] / Issue resolved: [detail]]

**Next Touchpoint:** [Date and type]

**Internal Notes (not customer-visible):**
[Anything the CSM observed that should not be in the customer-facing record: political dynamics, competitor mentions, relationship concerns, stakeholder observations]

What Makes This Different from Just Writing Notes

The structured format ensures:

  1. Outcome classification drives downstream workflows automatically -- a "concern surfaced" outcome can trigger a risk review without the CSM manually flagging it
  2. Sentiment tracking over time reveals trends invisible in individual meetings. A customer who moved from "enthusiastic" to "neutral" to "concerned" over three meetings is on a trajectory, even if each individual meeting seemed fine
  3. Action items with owners and deadlines prevent the "we discussed it but nobody did it" pattern
  4. Internal notes separate what the customer should see from what the CSM observed -- political dynamics, competitive mentions, and relationship assessments do not belong in a customer-visible record
  5. Consistency makes reporting meaningful. If every CSM logs meetings the same way, portfolio-level reporting (engagement health, outcome distribution, sentiment trends) becomes reliable

Quality Gates

  • Is the outcome classification honest? "Positive progress" should mean actual progress, not just a pleasant conversation. If nothing advanced, it is "status quo" -- and that is fine
  • Does the sentiment match what you observed, not what you hope? Optimistic sentiment logging is the most common data quality problem in CS. "Satisfied" should mean they expressed satisfaction, not that they did not complain
  • Are action items specific with owners and deadlines? "Follow up" is not an action item. "Send API documentation to Lisa Park by Friday March 14" is
  • Are internal notes genuinely useful? If you would not want your manager or successor to read them, they should not be there. If they contain context that would help your manager or successor, they should be there
  • Would this record allow another CSM to understand what happened without asking you? That is the bar. If your notes are only useful to you, they are not CRM-ready

Principles

  • Log the meeting within 1 hour while the memory is fresh. Meeting records written the next day lose nuance and accuracy. Meeting records written a week later are fiction
  • The outcome and sentiment classifications are more valuable than the narrative. Narratives are read once. Classifications aggregate into portfolio-level data that reveals patterns across hundreds of meetings
  • Every no-show is a data point. Do not skip logging no-shows. They are engagement signals that compound -- one no-show is noise, three no-shows is a pattern, five no-shows is a risk
  • Internal notes are for context, not for venting. "Customer was rude" is venting. "Customer was visibly frustrated about the unresolved API issue -- approached the conversation defensively and was not receptive to the adoption discussion I planned" is context that helps the next person engage effectively

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 326,506. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.