log-call-transcripts

log-call-transcripts is a skill for Claude Code, Codex from microsoft/dataverse-business-skills. It costs 76 tokens per session (2,781 once invoked), scanned A, original, MIT.

A sales-recording helper that reads a sales call transcript and creates a Phone Call activity in Dataverse, Microsoft's business-data platform. It extracts people, companies, call details, and useful discussion points from the transcript.

In plain words
What is it for?
Use it to process a transcript, identify participants and call metadata, summarize the topic, and save the resulting activity to the CRM.
Why use it?
It removes the need to enter call information into the CRM by hand after every conversation. This helps keep customer records more complete and consistent.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

About the project

Dataverse Business Skills is a collection of natural-language instructions that teach AI agents how to follow business processes, policies, and domain knowledge for Microsoft Dataverse. Organizations use the skills with Dataverse environments connected to products such as Power Apps, Dynamics 365, or Power Platform. The catalogue entries are skills from this collection.

microsoft/dataverse-business-skills · 49 stars · on GitHub

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/microsoft/dataverse-business-skills/log-call-transcripts
Any agent
npx skills add microsoft/dataverse-business-skills --skill log-call-transcripts
Clone the repo
git clone --depth 1 https://github.com/microsoft/dataverse-business-skills

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for log-call-transcripts

README.md
[![agentmods](https://agentmods.dev/badge/skills/microsoft/dataverse-business-skills/log-call-transcripts.svg)](https://agentmods.dev/skills/microsoft/dataverse-business-skills/log-call-transcripts)
Your own site
<a href="https://agentmods.dev/skills/microsoft/dataverse-business-skills/log-call-transcripts"><img src="https://agentmods.dev/badge/skills/microsoft/dataverse-business-skills/log-call-transcripts.svg" alt="Measured on agentmods" height="20"></a>
Per session 76 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,781 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00076 $0.02781
Opus 5 $0.00038 $0.01391
Sonnet 5 $0.00015 $0.00556
Haiku 4.5 $0.00008 $0.00278

Measured 6d ago against content hash d5f4398370f4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

log-call-transcripts scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/log-call-transcripts/SKILL.md · 294 lines

How it starts

The opening of the file, as written. The whole thing — 294 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Log Call Transcripts

When a sales representative uploads or provides a call transcript, this skill extracts actionable information from the conversation and creates a properly structured Phone Call activity record in Dataverse. This eliminates manual data entry and ensures consistent activity logging across the sales team.

Instructions

Step 1: Receive and Parse Transcript

When the user provides a call transcript:

  1. Accept the transcript content (text or file upload)
  2. Identify the transcript format and parse it appropriately
  3. Extract speaker labels if available (e.g., "Sales Rep:", "Customer:")
Step 2: Extract Key Information from Transcript

Analyze the transcript to identify and extract:

Participant Information:

  • Customer name(s) mentioned
  • Company/Account name mentioned
  • Customer's role or title if mentioned
  • Phone numbers discussed

Call Metadata:

  • Call duration (if mentioned or calculable)
  • Date/time of call (if mentioned, otherwise use current datetime)
  • Call direction (inbound/outbound based on context)

Content Analysis:

  • Subject/Topic: Main purpose of the call (e.g., "Product demo follow-up", "Pricing discussion", "Support inquiry")
  • Key Discussion Points: Summarize 3-5 main topics discussed
  • Action Items: Any commitments or next steps mentioned
  • Customer Pain Points: Problems or challenges the customer mentioned
  • Competitor Mentions: Any competitor products or companies referenced
  • Budget Signals: Any pricing or budget discussions
  • Decision Maker Signals: References to decision-making process or stakeholders
  • Timeline Signals: Any urgency or timing requirements mentioned
Step 3: Match to Existing Records

Query Dataverse to find matching records:

Find Account:

Use read_query to search the account table:
- Search by company name mentioned in transcript
- Match on account.name field
- If multiple matches, list them for user confirmation

Find Contact:

Use read_query to search the contact table:
- Search by participant name(s) from transcript
- Match on contact.fullname or contact.firstname + contact.lastname
- Filter by accountid if account was identified
- If multiple matches, list them for user confirmation

Read the full file on GitHub · 294 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 294 lines · 76 tokens per session scan A d5f4398370f4

Subscribe to this mod's changes

log-call-transcripts is a skill published in the GitHub repository microsoft/dataverse-business-skills (49 stars, last pushed 5mo ago), licensed MIT. It adds 76 tokens to every session and 2,781 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.