Manage priorities
Manage a priorities tracker in the configured knowledge base tool with canonical Table/Board views, status-based cards, and per-priority detail subpages linked from a Link property. Use when the user asks to manage priorities, update top priorities, add priority cards, create priority detail subpages, or bootstrap the tracker.From its SKILL.md
npx -y skills add vecten/sdlc-toolkit --skill manage-prioritiesAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- reads credentialsReads from 1 credential source: `knowledge_base_tool`.
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
5.7 KB, ~1.2k tokens by cl100k_base, as published. Nobody here has run it
Manage Priorities
Use this workflow to manage a priorities board in the configured knowledge base tool.
This skill reads knowledge_base_tool from config. Currently supported: notion.
Primary outcomes:
- keep a clean priorities database with
StatusandLink - keep two canonical views:
TableandBoard - add and update priorities quickly
- create a detail subpage per priority and store its URL in
Link
Config dependency
Read <workspace>/.cursor/skills-config.yaml if available. This skill works without config but benefits from it.
Optional config keys:
notion.priorities_statuses— custom status optionsnotion.priorities_database_id— cached database ID to skip discoverynotion.custom_input— additional properties, custom view names
Guardrails
- Ask for clarification when target page/database is ambiguous.
- Do not perform destructive schema/view cleanup unless the user explicitly asks.
- Keep responses concise and include resulting URLs.
Required Inputs
Collect from the user prompt:
- target scope (teamspace/page/database), if provided
- requested operation:
- maintain existing tracker
- add/update priorities
- create subpages and link them
- bootstrap tracker (only when explicitly requested)
- priority text and desired status when adding/updating items
If critical inputs are missing, ask using structured choices when possible.
Canonical Tracker Definition
Data source properties
Use these canonical properties:
Priority(title property)Status(select): read options fromnotion.priorities_statusesin config, or default toOn Track,Planned,At Risk,Blocked,DoneLink(URL): detail subpage URL for the priority
If equivalent properties exist with different names, prefer renaming to canonical names after confirmation.
Canonical views
Table- visible properties:
Priority,Status,Link - default sort:
Statusascending
- visible properties:
Board- grouped by:
Status - card properties:
Priority(and optionalLink)
- grouped by:
If one or both are missing, create them. If they exist, update them.
Read notion.custom_input for any custom view names or additional properties.
Runtime Discovery Flow
Step 1 — Locate target workspace context
- If
notion.priorities_database_idis set in config, fetch that database directly. - If the user provided a page/database URL or ID, fetch it directly.
- If not, search for likely pages (for example: "Top Priorities", "Current Priorities").
- If multiple plausible targets are found, ask the user to pick one.
Step 2 — Resolve database/data source
- Fetch the selected page or database.
- If the page contains more than one priorities-like database, ask the user which one to use.
- Capture: page ID (parent), database ID, data source ID, existing views and schema.
- If
notion.priorities_database_idis null in config and config exists, cache the discovered ID back into config.
Step 3 — Normalize schema and views
Apply canonical schema and views carefully:
- add missing properties (
Status,Link) without deleting unknown columns - enforce
Statusoptions from config or defaults - create/update
TableandBoard
Only remove extra properties/views when the user asks for cleanup.
Operation: Maintain Existing Tracker (Default)
When the user asks to "manage priorities" without bootstrap intent:
- Run runtime discovery.
- Ensure canonical schema.
- Ensure
TableandBoardviews. - Report concise summary.
Operation: Bootstrap Tracker (Explicit Opt-In)
Run only when user explicitly asks to create/setup/bootstrap a tracker.
- Pick target parent page/teamspace (ask if unclear).
- Create page (for example,
Current Priorities) if missing. - Create database under that page with canonical properties.
- Create/update canonical views.
- Return page URL, database URL, and data source ID.
Operation: Add Priority
When user provides new priorities:
- Ensure tracker is normalized (schema/views).
- Create one database row per priority with:
Priority: provided textStatus: provided value (default toPlannedif unspecified)- optional icon if provided
- For each new row, create a detail subpage and write its URL to
Link. - Return created items with their row URLs and detail page URLs.
Operation: Update Priority
When user asks to update existing priorities:
- Locate target item(s) by exact title first.
- If multiple items match, ask user to disambiguate.
- Update requested fields.
- If
Linkis empty and user wants detail pages, create subpage and store URL. - Return list of updated items with final status and links.
Subpage + Link Workflow
Use this flow whenever the request includes "details", "subpage", or "link property".
- For each priority row, create a page under the row page (preferred) or parent tracker page (fallback).
- Use title pattern:
<Priority> - Details - Add minimal content scaffold:
## Context
## Notes
## Related Links
- Save the subpage URL into the row's
Linkproperty. - If an existing
Linkis stale or blank, replace it after confirming intent.
Output Format
Updated priorities tracker.
- Page: <title> (<url>)
- Database: <title> (<url>)
- Views: Table, Board
- Changes:
- <change 1>
- <change 2>
Priority items:
- <priority title> — Status: <status> — Link: <url or None>
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.