Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
DeepSeek
Muse+22 moreClaude CodeCursorCodex CLIDeepSeekMuse
Dots
Manus
Grok
Grok Bot+18 agentsAction Item Recovery
Recover real commitments, owners, deadlines, and blockers from meetings, chats, emails, and notes.
$5
Action Item Recovery
Example session with this skill installed
Recover action items from notes dated September 14, 2026 UTC: Maya asked Lina to finish payment retry tests by September 18, and Lina replied “I’ll own it.” Omar suggested updating the rollout guide, but nobody accepted. The team needs staging credentials before verification; no owner or deadline was named.
- Read your context and instructions
- Compiled the action item recovery
Action recovery report
Confirmed commitments
| Action | Classification | Owner | Due date | Evidence |
|---|---|---|---|---|
| Finish payment retry tests | Committed | Lina | Sep 18, 2026 | Maya requested it and Lina explicitly said, “I’ll own it.” |
Requested or suggested—not yet committed
| Item | Status | What is missing |
|---|---|---|
| Update the rollout guide | Suggested | An accepting owner and deadline |
| Provide staging credentials | Required dependency | Owner, delivery date, and confirmation |
Dependency chain
Verification cannot begin until staging credentials are available. The source does not identify who owns that dependency, so no owner was inferred.
Follow-up draft
“Quick ownership check: Lina has accepted payment retry testing for September 18. We still need an owner and delivery date for staging credentials before verification can start. Omar’s rollout-guide idea also needs an accepting owner and deadline if we want it tracked as work.”
Audit notes
- One explicit commitment recovered.
- One suggestion kept separate from committed work.
- One unassigned dependency flagged.
- No deadline or owner was invented.
Connects securely to your tools. The creator never sees your data.
About this skill
The problem
Meeting transcripts and chat logs often bury actual commitments under layers of context, leading to missed deadlines and "ownerless" tasks. General summaries frequently fail to distinguish between a casual suggestion and a binding project commitment.
What it does
- Extracts concrete tasks, owners, and deadlines from messy unstructured text like Zoom transcripts or Slack exports.
- Classifies statements into committed, requested, or suggested actions to prevent "wishful thinking" from becoming tasks.
- Identifies ownership conflicts and unassigned responsibilities that need immediate human resolution.
- Maps dependencies and required approvals to show the ripple effect of delayed items.
- Generates a neutral follow-up draft to re-engage owners without manual drafting.
Why this beats prompting it yourself
Generic LLM summaries often hallucinate owners or miss subtle relative deadlines. This skill uses a strict evidence-based classification system to ensure every item in the tracker is backed by specific fragments from the source, preventing the creation of "phantom" tasks.
Use cases
- Turning a messy Zoom transcript into a structured Project Management board update.
- Auditing a long email thread to find who actually agreed to do what by when.
- Generating a list of unassigned blockers after a chaotic planning session.
- Tracking how deadlines shifted over the course of a multi-day chat discussion.
Known limitations
Relative deadlines (e.g., "by next Friday") are only normalized if the source date is provided. It will not infer owners based on proximity to a mention.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Business & Operations.
Requires explicit source dates for deadline normalization. Won't guess owners without clear evidence. Focuses on extraction, not task execution or management.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 1 month ago
Permissions
No special permissions declared or detected