✅ Meeting Action Item Verifier
Pull the real commitments out of a meeting transcript, with owners, and separate firm action items from tentative "let's revisit" ones. Reads .vtt, .srt, .txt, or pasted notes, attributes each item to the speaker who said it, classifies it firm or tentative, and marks any decision with no clear owner as needs-confirmation instead of inventing one.
- Separate firm commitments from tentative 'let's revisit' suggestions.
- Automatically attribute tasks to specific meeting participants.
- Identify ownerless decisions that require immediate follow-up.
$12
· or 60 creditsSecure checkout via Stripe
Included in download
- Separate firm commitments from tentative 'let's revisit' suggestions.
- Automatically attribute tasks to specific meeting participants.
- file_read, terminal automation included
- Ready for Cursor
Sample input
Find the real to-dos from this meeting transcript and identify who is responsible for each one.
Sample output
### Meeting Action Items - **Firm**: @Sarah to update the API documentation by Friday. - **Firm**: @James to fix the login bug in the staging environment. - **Tentative**: Explore migrating to AWS Lambda (to be revisited next month). - **Needs-Confirmation**: Implement SOC2 compliance audit (unattributed).
✅ Meeting Action Item Verifier
Pull the real commitments out of a meeting transcript, with owners, and separate firm action items from tentative "let's revisit" ones. Reads .vtt, .srt, .txt, or pasted notes, attributes each item to the speaker who said it, classifies it firm or tentative, and marks any decision with no clear owner as needs-confirmation instead of inventing one.
$12
· or 60 creditsSecure checkout via Stripe
Included in download
- Separate firm commitments from tentative 'let's revisit' suggestions.
- Automatically attribute tasks to specific meeting participants.
- file_read, terminal automation included
- Ready for Cursor
- Instant install
Sample input
Find the real to-dos from this meeting transcript and identify who is responsible for each one.
Sample output
### Meeting Action Items - **Firm**: @Sarah to update the API documentation by Friday. - **Firm**: @James to fix the login bug in the staging environment. - **Tentative**: Explore migrating to AWS Lambda (to be revisited next month). - **Needs-Confirmation**: Implement SOC2 compliance audit (unattributed).
About This Skill
Precision Transcript Extraction
Transform messy meeting transcripts into high-signal action plans. This skill solves the problem of "meeting amnesia" by distinguishing between firm commitments and tentative ideas that often get lost in the chatter. Unlike basic summarizers that provide vague bullet points, this tool rigorously classifies tasks based on speaker intent and owner attribution.
What it does
- Intelligent Classification: Categorizes items as Firm (committed), Tentative (to be revisited), or Needs-Confirmation.
- Owner Attribution: Automatically maps tasks to specific speakers based on transcript labels.
- Safety Guardrails: Prevents "artificial certainty" by marking unattributed decisions for human review rather than guessing owners.
- Structural Integrity: Outputs clean Markdown that identifies exactly what was promised and by whom.
Why use this skill?
Prompting an AI yourself often leads to "hallucinated commitments" where the AI turns a casual suggestion into a hard task. This skill uses a specialized extraction workflow that respects the nuance of spoken language. It is optimized for developers and project managers who need reliable, audit-ready to-do lists from internal syncs or client calls.
Use Cases
- Separate firm commitments from tentative 'let's revisit' suggestions.
- Automatically attribute tasks to specific meeting participants.
- Identify ownerless decisions that require immediate follow-up.
- Create structured Markdown to-do lists from raw transcript files.
Known Limitations
It is an extractor, not a notetaker. It finds and classifies candidate commitments by cue phrases, so an unusually worded commitment can be missed and a strongly worded non-commitment can surface. It never fabricates an owner or hardens tentative language into a deadline; items it cannot confidently attribute are flagged for you to confirm.
How to Install
mkdir -p ~/.claude/skills && curl -sL https://www.agensi.io/api/install/meeting-action-item-verifier -o /tmp/meeting-action-item-verifier.zip && unzip -o /tmp/meeting-action-item-verifier.zip -d ~/.claude/skills && rm /tmp/meeting-action-item-verifier.zipFree skills install directly. Paid skills require purchase - use the download button above after buying.
Reviews
No reviews yet - be the first to share your experience.
Only users who have downloaded or purchased this skill can leave a review.
Early access skill
Be the first to review this skill.
Only users who have downloaded or purchased this skill can leave a review.
Security Scanned
Passed automated security review
Permissions
File Scopes
Read-only. Also accepts a pasted transcript via --stdin. Prints a structured list (classification, owner, item). It attributes an owner from the speaker labels in the transcript; without labels it returns "unattributed, confirm." Writes nothing and reads no environment variables.
Works with any agent that can read a transcript and run a local Python script (Claude Code, Cursor, Codex CLI, and other SKILL.md-compatible agents). Standard library only, no install step. Owner attribution works best on a speaker-labeled transcript.