todoist-browser-tasks
Skill9 starsChanged 37 days agoMITView on GitHub
Manage Todoist tasks through the teammate's signed-in browser: list, create, correct and complete with readback verification.
In this repository13 other files it gives its agents.
- site/llms-full.txtllms.txt
- site/llms.txtllms.txt
- checked-spreadsheet-analysisSkill
- document-to-action-itemsSkill
- gmail-browser-draftSkill
- google-calendar-browser-tasksSkill
- google-drive-browser-tasksSkill
- grounded-researchSkill
- meeting-action-itemsSkill
- private-team-reviewSkill
- systematic-debuggingSkill
- weekly-review-planningSkill
- use-mac-appsSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 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.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use the teammate's own persistent browser profile at app.todoist.com.
- Distinctive: line 12
Create with Add task: fill "Task name" and submit with Enter.
- Distinctive: line 16
Delete through the task dialog's More actions menu, Delete item, and the "Delete task?" confirmation naming the exact title.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Todoist in the teammate browser
Use the teammate's own persistent browser profile at app.todoist.com. Every write below still needs its normal exact approval first. Single-session discipline: drive the bot's own browser session only, never a parallel profile copy.
- Open the Inbox and verify the signed-in account matches the connected owner before acting. On any login wall, expired session, 2FA, CAPTCHA or unexpected account, stop and request private sign-in. Never work around it, and never copy the login to another teammate.
- List tasks by reading the Inbox, Today or Upcoming rows. Address a task by its exact shown title, never by row position alone. Re-list after every mutation; the list is the readback.
- Create with Add task: fill "Task name" and submit with Enter. The quick-add form has no description field. Read the new row back in the list before claiming success; a toast alone is not proof.
- Correct through the task's detail dialog: "Task name" for titles, "Task description" (textbox, then Save) for descriptions, "Date" (quick-pick or typed date) for due dates, Priority listbox ("Priority 1" urgent through "Priority 4" normal). Save, then read the dialog back.
- Complete with the row's "Mark task as complete" checkbox (or the dialog checkbox). The task must disappear from Inbox and Today. A missing readback is uncertainty, never success; never repeat a close you cannot confirm.
- Reopen through the activity log: project options menu, View activity, follow the "Open task" link of the completed entry, uncheck the completed checkbox in its dialog. The task must reappear in the Inbox list. The immediate Undo toast is not a reopen proof.
- Delete through the task dialog's More actions menu, Delete item, and the "Delete task?" confirmation naming the exact title. The task must disappear from the Inbox and stay absent. Deletion is permanent: only delete approved private test tasks, never anything else.
- State page, account, task title and exact field change in each proposal. Touch only approved tasks. Never extract or reuse session cookies or tokens.
Provenance
Sidemates maintained built-in method. Proven in a signed-in teammate browser: create, edit, complete, reopen and delete, each read back, inbox restored. Instructions grant no tool or account permissions.
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.

