miseenplace: turn chat asks into your to-do list

    1

    Turns the asks in your chat into an action list and drafts replies it never sends without your yes.

    Free

    1 useSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    miseenplace: turn chat asks into your to-do list

    Example session with this skill installed

    Check my messages and tell me what I owe people.

    • Read your context and instructions
    • Compiled the miseenplace

    Real output from mep demo (fictional data).

    1. Only what needs you

    $ mep inbox
    3 · Rowan Hale · DM: Rowan Hale · Could you review the parser PR when you get a chance? It blocks the beta. · todo
    4 · Sam Okafor · #platform > incident follow… · @Jordan Rivers can you write up the retro for yesterday's outage by Friday? · todo
    5 · Alex Chen · #platform > release 2.4 · Changelog still pending, ETA? · reply
    

    Message 1 asked who owns the changelog. You had already answered in that thread, so it is not listed.

    2. Bookkeeping, one call each

    $ mep triage 3 --todo
    todo #1
    $ mep triage 4 --todo
    todo #2
    

    3. A reply is a draft first. Nothing is posted.

    $ mep triage 5 --reply-draft 'sending it Thursday'
    draft dc5ceb5aa -> #platform > release 2.4
    ----- exact text -----
    > Changelog still pending, ETA?
    
    sending it Thursday
    ----- end -----
    A human reads this exact text. Only an explicit yes on it justifies the next step.
    

    4. Free text can never be sent

    $ mep send 'hello'
    miseenplace: 'hello' is not a draft id (looks like d1a2b3c4e); send takes only the id printed by `draft`, never message text
    $ mep send dc5ceb5aa
    posted 7
    $ mep send dc5ceb5aa
    miseenplace: draft dc5ceb5aa was already sent (ref 7)
    

    5. What to do first, and the stand-up written from the log

    $ mep top && mep standup --since today
    - **#1** Could you review the parser PR when you get a chance? It blocks the beta. · todo · `DM: Rowan Hale` · Rowan Hale
    - **#2** @Jordan Rivers can you write up the retro for yesterday's outage by Friday? · todo · `incident follow-ups` · Sam Okafor
    **Did** (since 2026-10-23):
    - triaged the inbox with miseenplace — 10m
    
    **Next:**
    - Could you review the parser PR when you get a chance? It blocks the beta.
    - @Jordan Rivers can you write up the retro for yesterday's outage by Friday?
    

    Summary: three asks handled. Two are journal items in priority order, and one reply went out only after your yes. Sending it twice was refused.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Chat is where work gets asked for and promised, and it scrolls away. To-do apps start empty and trust you to copy things in. miseenplace reads your chat, keeps only what needs you, and turns it into an audit-logged action list your agent drives: a chef's mise en place for your workday.

    Six skills in one package

    • Setup: installs the CLI, connects your chat and verifies it.
    • Inbox Triage: one line per actionable ask (who, where, the ask, a suggested action); each becomes a journal item in one call. Asks you already answered in the thread are hidden.
    • Inbox Watch: waits in the background and wakes your agent only when someone needs you.
    • Journal: action items, waiting-on items, a work diary and a stand-up written from the log.
    • End-of-Day Review: closes what is really done, lists what is still on you, and saves lessons only from your real corrections.
    • Learn My Style: learns how you write from your own sent messages, so drafts sound like you.

    Safe by design

    • Replies are stored as drafts. Nothing is posted until you say yes to that exact text.
    • send takes only a draft id, so an agent cannot post free text.
    • Chat content is treated as data, never as instructions. Credentials are never printed.

    Cheap to run The CLI does the classification and bookkeeping, so a small model can drive the loop. A ten-message triage stays near 500 tokens of output.

    Requirements Requires the free, open-source miseenplace CLI (Apache-2.0): uv tool install miseenplace-cli. Source and docs: https://github.com/ankitkpandey1/miseenplace. The chat source today is Zulip, and mep demo gives a two-minute tour on fictional data with no account. Works with Claude Code, Codex, Gemini CLI and Cursor.

    Needs the miseenplace CLI installed. The live chat source today is Zulip (mep demo runs without an account). Replies are never sent as free text: every message goes through a stored draft and your explicit yes.

    How to use

    1. 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. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    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

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 25 days ago

    • Free with an Agensi account

    Requires

    miseenplace CLI

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access
    Environment Variables
    Listed25 days ago

    What's inside

    Frequently Asked Questions

    The free, open-source miseenplace CLI (`uv tool install miseenplace-cli`, Python 3.11+) and, for live use, a Zulip account with a personal API key. The Setup skill walks your agent through it; `mep demo` works with no account at all.

    No. Replies are stored as drafts and shown to you as exact text. `mep send` takes only a draft id, never free text, and your agent runs it only after your explicit yes to that text. Sending the same draft twice is refused.

    The Learn My Style skill fetches your own sent messages (`mep style learn`), your agent writes a short style file with real quotes, and you correct it. Every draft reads that file first. Only your own messages are used.

    Each ask gets one action: todo (a journal item), waiting (with a nudge date), reply (a draft), or ignore. Asks you already answered in the thread never show up.

    Seven SKILL.md files: a top-level miseenplace skill that routes to the six skills Setup, Inbox Triage, Inbox Watch, Journal, End-of-Day Review and Learn My Style, plus their reference notes. The same skills ship free and open source with the CLI.

    New in Workflow Automation