ainote
Skill0 starsChanged 25 days agoMITView on GitHub
Use this skill first when the user wants to operate real Office Notes / Read-Write Notes data, including viewing, generating, creating, updating, moving, completing, or deleting real notes, folders, schedules, reminders, todos, or tasks. Note types include writing, records, mind maps, and text notes, so classify intent carefully. Cover natural phrases such as creating a note, writing a note, and saving content to Read-Write Notes. Do not use this skill only when the user is clearly generating or editing a document, operating local files, working with external app data, or the request is clearly not about real Office Notes / Read-Write Notes data.
In this repository60 other files it gives its agents.
- 21st-aiSkill
- 21st-cli-useSkill
- 21st-design-syncSkill
- 21st-registrySkill
- 21st-ui-buildSkill
- 21st-ui-exploreSkill
- 21st-ui-reviewSkill
- ab-testingSkill
- ad-creativeSkill
- adsSkill
- ai-seoSkill
- analyticsSkill
- asoSkill
- attributionSkill
- cavemanSkill
- churn-preventionSkill
- cold-emailSkill
- co-marketingSkill
- community-marketingSkill
- competitor-profilingSkill
- competitorsSkill
- content-strategySkill
- copy-editingSkill
- copywritingSkill
- croSkill
- customer-researchSkill
- directory-submissionsSkill
- emailsSkill
- eventsSkill
- free-toolsSkill
- imageSkill
- influencer-marketingSkill
- launchSkill
- lead-magnetsSkill
- marketing-councilSkill
- marketing-ideasSkill
- marketing-loopsSkill
- marketing-planSkill
- marketing-psychologySkill
- meeting-assistantSkill
- offersSkill
- onboardingSkill
- paywallsSkill
- pedrojaimot-file-organizerSkill
- popupsSkill
- pricingSkill
- product-marketingSkill
- programmatic-seoSkill
- prospectingSkill
- public-relationsSkill
- referralsSkill
- revopsSkill
- sales-enablementSkill
- schemaSkill
- seo-auditSkill
- signupSkill
- site-architectureSkill
- smsSkill
- socialSkill
- videoSkill
The file
We split this skill into the rules it gives an agent
11 rules. About 0 are common, 3 are distinctive, and 8 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
8
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 14
When the user's goal involves real notes, schedules, or tasks, route to ainote first, then dispatch to the appropriate module.
- Very specific: line 18
If a single request contains both a time expression and an action or item, route to schedule by default.
- Very specific: line 23
Prefer note only when the user explicitly asks to write a note, record a note, write meeting minutes, generate a summary, or record the …
- Very specific: line 39
Write operations and sync: shared/write-and-sync.md
- Very specific: line 43
Mandatory, no exceptions: all Ainote API calls must be sent through shared/scripts/ainote_api.py.
- Very specific: line 44
Mandatory, no exceptions: do not call python bare. Prefer the AINOTE_PYTHON environment variable.
- Distinctive: line 53
Check the module health endpoint for the requested local OpenModel API.
- Distinctive: line 54
If the local service is unavailable, try to start the installed AINOTE app with its background OpenModel argument.
- Distinctive: line 55
Wait briefly for the local service to become ready.
- Very specific: line 111
Do not expose API identifiers in user-facing answers, including id, uuid, localId, noteId, sourceId, pageId, taskId, and any *Id / *ID …
- Very specific: line 112
Do not generate links that contain API identifiers, including ainote://...
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
ainote
Unified Ainote skill. Currently supported modules: note and schedule.
Overview
This skill is the unified entry point for real data operations in Ainote / Read-Write Notes.
When the user's goal involves real notes, schedules, or tasks, route to ainote first, then dispatch to the appropriate module.
Routing Priority
- If a single request contains both a time expression and an action or item, route to
scheduleby default. - Time expressions include, but are not limited to:
today,tomorrow,the day after tomorrow,tonight,tomorrow morning,afternoon,9am,3:30,next Monday,before Friday, andend of month. - Actions or items include, but are not limited to:
meeting,appointment,remind,schedule,move to,postpone,move earlier,complete,submit, andhandle. - Even if the user does not explicitly say
schedule,task,reminder,event, ortask, do not prefernotewhen the semantics are time-based. - Route
关注星,focus star,starred snippet, ornimblequeries toschedule. - Prefer
noteonly when the user explicitly asks to write a note, record a note, write meeting minutes, generate a summary, or record the given content.
Environment
Runtime dependencies are listed in
meta.json.
Module Routing
| User intent | Module | Read |
| Search notes, read content, create/move/rename/delete notes or folders | note | note/SKILL.md |
| View schedules, create/update/delete events, reminders, todos, or tasks, query focus stars, and handle any time-based natural-language scheduling request such as "schedule a standup tomorrow at 9am" | schedule | schedule/SKILL.md |
Shared Rules
- Port and health checks:
shared/port-and-health.md - Write operations and sync:
shared/write-and-sync.md
Network Request Rules
Mandatory, no exceptions: all Ainote API calls must be sent through shared/scripts/ainote_api.py. Do not call Invoke-RestMethod or curl directly.
Mandatory, no exceptions: do not call python bare. Prefer the AINOTE_PYTHON environment variable; if it is empty, use python3 on macOS / Linux / zsh / bash.
Reason: PowerShell 5.1 silently corrupts Chinese text in two places: request bodies may be re-encoded from UTF-8 to ANSI, and command-line arguments passed to external processes may be affected the same way. Therefore,
--body '...'string arguments are unsafe on Windows and a file or encoded body must be used instead.
Local Service Startup Flow
shared/scripts/ainote_api.py is the only startup-aware request entry. Every request through this script follows this flow:
- Resolve the current OpenModel port from the AINOTE app config, then fall back to
46588. - Check the module health endpoint for the requested local OpenModel API.
- If the local service is unavailable, try to start the installed AINOTE app with its background OpenModel argument.
- Wait briefly for the local service to become ready.
- Send the original request.
Do not manually start AINOTE with shell commands before API calls. Do not repeat the health-check sequence yourself. If the script cannot find or start AINOTE, report that the AINOTE local service is unavailable or that the user may need to open and sign in to AINOTE.
Background startup depends on AINOTE supporting its background OpenModel launch argument. When AINOTE is already running, the app must treat that argument as a service wake-up and must not show the main window.
GET Requests
PYTHON_BIN="${AINOTE_PYTHON:-python3}"
"$PYTHON_BIN" ".ainote/skills/ainote/shared/scripts/ainote_api.py" \
--url "http://127.0.0.1:46588/open-model-schedule/health" \
--token "$TOKEN"POST / PUT Requests (must use --body-encoded)
# 1. Build JSON
PYTHON_BIN="${AINOTE_PYTHON:-python3}"
body='{"type":"event","title":"Title","startTime":1234567890}'
# 2. URL percent-encode the body. The output is pure ASCII and bypasses encoding layers.
encodedBody=$("$PYTHON_BIN" -c 'import sys, urllib.parse; print(urllib.parse.quote(sys.argv[1], safe=""))' "$body")
# 3. Pass the encoded body to the script.
"$PYTHON_BIN" ".ainote/skills/ainote/shared/scripts/ainote_api.py" \
--url "http://127.0.0.1:46588/open-model-schedule/create" \
--method POST \
--body-encoded "$encodedBody" \
--token "$TOKEN"Why URL encoding is safest: URL percent-encoding turns all non-ASCII characters into ASCII sequences such as
%E6%99%A8. ASCII is unaffected by encoding layers. Python then usesunquoteto restore the correct Unicode string before sending UTF-8 bytes.Never do this (Chinese text will be corrupted):
python ainote_api.py --body '{"title":"Title"}' ...
Response Style
Default response structure:
ResultMatchesAdditional NotesNext Step
High-risk action confirmation structure:
Target ObjectIdentification EvidenceRisk NoticeAwaiting Confirmation
Special formatting requirements:
- Do not expose API identifiers in user-facing answers, including
id,uuid,localId,noteId,sourceId,pageId,taskId, and any*Id/*IDfield. - Do not generate links that contain API identifiers, including
ainote://...links. pageIdis an internal page locator, not a user-facing page number. Do not display it as页/page.
Examples
- "Search for my weekly report notes and read the content" -> enter
note/SKILL.md - "Summarize this week's note highlights" -> enter
note/SKILL.md - "Create a reminder for tomorrow at 2pm" -> enter
schedule/SKILL.md - "Standup tomorrow morning at 9am" -> enter
schedule/SKILL.md - "Remind me to submit the weekly report tonight at 8pm" -> enter
schedule/SKILL.md - "Add a todo due before Friday to submit the proposal" -> enter
schedule/SKILL.md - "Move this afternoon's meeting to 4pm" -> enter
schedule/SKILL.md
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

