More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+22 more

    Carrynote: pick up where the last session stopped

    1

    Saves dated notes, plan progress and open items on request, resumes with a short done, next, blocked report, and writes cross-repo handoff files with status, data contract and a verify command.

    $9

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Carrynote: pick up where the last session stopped

    Carrynote: pick up where the last session stopped

    Example session with this skill installed

    Resume. Where did we leave off on the search feature?

    • Read your context and instructions
    • Compiled the carrynote

    Read notes/index.md, plans/search.md PROGRESS, git status (clean) and git log -10.

    Resume: search feature

    • Done (6 of 9): items 1 to 6 shipped. Last commit a1b2c3d "search: debounce input".
    • Next: item 7, highlight matched terms in results.

    Blocked (1): item 8 waits on the API repo. Handoff handoffs/2026-10-17-search-index.md in api-service is still STATUS: OPEN.

    • Open items: rotate the staging search key (noted 2026-10-15).

    Re-verified: note from 2026-09-10 said "fuzzy match deferred". Checked the code: still not implemented, so the note stands.

    That accounts for 6 done, 1 next, 1 blocked and 1 still pending (item 9, empty-state copy).

    Handoff on file (api-service)

    STATUS     OPEN
FROM       web-app session 2026-10-17
TASK       add search_vector column to products · GIN index · expose /search?q=
CONTRACT   products.search_vector tsvector · GET /search returns {id, name, rank}
RESOURCES  DB port 54422
VERIFY     curl -s "localhost:4000/search?q=mug" | jq length

    Starting item 7 now.

    Next steps

    • Relay one line to the api-service session: "read handoffs/2026-10-17-search-index.md".
    • Rotate the staging search key and tell me, so I can close the open item.
    • Say "save" when you stop and I will update PROGRESS and the notes.

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

    About this skill

    Every new session starts from zero: your agent does not know what was decided yesterday, the plan restarts, and the other repo never got the task it was waiting on. You end up re-explaining the same context and acting as the message bus between sessions.

    Carrynote keeps work continuous across sessions, context compaction and repos, using plain dated notes, a plan PROGRESS section and handoff files.

    What it does

    • Session brief on boot: last recap, open items, active project and at most 3 attention flags, 12 lines max.
    • Save on request routes each fact to the note it belongs in, updates plan PROGRESS and an open-items list, and reports what was saved in three lines. Only on request, never silently.
    • Resume reads the notes index, plan PROGRESS, git status and git log -10, re-checks claims older than about 30 days, and reports done, next and blocked in five lines.
    • Recall searches dated notes by keyword and cites them. If nothing is found it says so.
    • Cross-repo handoff writes handoffs/<date>-<slug>.md in the target repo with STATUS, FROM, TASK, CONTRACT, RESOURCES and VERIFY.
    • Connector check proves each needed MCP server with one cheap read call.

    How it works

    1. Boot the shared rules from one global file and give the brief.
    2. Save dated notes with last_verified, index lines, PROGRESS and open items.
    3. Resume from notes plus git, then continue the next item.
    4. Hand off with a data contract; the target marks DONE with evidence and the origin runs VERIFY.

    Frameworks & tools

    Works with plain Markdown notes and git. Gives Claude Code specifics (SessionStart hook, /mcp, SendMessage) and equivalents for other agents.

    Why this beats prompting it yourself

    "Remember this" prompts dump everything into one file that bloats and goes stale. Carrynote routes facts by topic, caps note sizes, archives old entries, and refuses to invent past decisions that no note records.

    Use cases

    • End of day "save everything, continue later".
    • Monday morning "where did we leave off?"
    • Two repos where a column rename in one must reach the other.
    • After compaction when the agent has lost its rules.

    Known limitations

    Not for one-off questions in a single session or for planning new work. Commits notes only when you allow it and never pushes without an explicit request.

    How to use

    Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill passes our trust checks below.

    Security scanned

    Passed 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 scanned

    Verified clean 1 day ago

    • Passed all security checks, Safe to install

    Requires

    git

    Listed1 day ago

    What's inside

    Frequently Asked Questions