agentsclimarketplace

Notion ao ops

Skill rooftop-Owl/notion-ao-research/skills/notion-ao-ops

Agent-oriented research workflow skills for Notion MCP. NL database operations, workspace setup, design handbook (EN/KO). Works with Claude Code, Cursor, Windsurf, or any MCP client.

Install
npx -y skills add rooftop-Owl/notion-ao-research --skill notion-ao-ops

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 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.

What its author says it does

Copied from the file, not written here

This skill should be used when the user wants to "add entry", "log work", "query schedule", "add milestone", "log experiment", "add paper", "mark complete", "what's on today", "show experiments", "update status", "session handoff", "end session", "write handoff", "pass to local", or any other natural language operation on research tracking databases in Notion. Provides NL→database routing, API patterns, and operation conventions for any research workspace.

The file declares its own license as CC-BY-NC-SA-4.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

5.5 KB, as published. Nobody here has run it

Notion Research

Purpose

Provide a natural-language interface for research-tracking databases through Notion MCP tools. Resolve operations at runtime from workspace configuration, classify user intent into reusable operation patterns, and dispatch to the correct reference guide for implementation details in any configured research workspace.

Runtime Resolution

  1. Read workspace configuration first.
  2. Match the user request to a database pattern by schema characteristics and intent cues.
  3. Extract runtime values from config: data_source_id, title property name, and allowed select/multi-select options.
  4. Build API calls using those runtime values only.
  5. If workspace configuration is missing, load the notion-ao-workspace skill and establish config before executing ADD/QUERY/UPDATE operations.

Database Operation Patterns

PatternCharacteristicsRoute to
Time-series logDate-indexed entries, category tags, effort/hours trackingdiary-interface.md
Status-tracked itemsLifecycle statuses, due dates, milestone/task-likemulti-db-interface.md
Experiment registryPhase workflow (designed→running→complete), findings, blockersmulti-db-interface.md
Reference-reading catalogAuthors, DOI/URL, reading status, relevancemulti-db-interface.md
Manuscript stage trackerDraft stage progression, deadlines, progress %multi-db-interface.md
Note archiveCategory-tagged, date-stamped, free-form textmulti-db-interface.md

Matching signals

  • Time-series log: add entry, log work, what's on today, this week, log hours
  • Status-tracked items: add milestone, mark complete, upcoming, at risk
  • Experiment registry: log experiment, set blocker, log findings, set phase
  • Reference-reading catalog: add paper, what am I reading, add citation
  • Manuscript stage tracker: add chapter, move to drafting, progress, deadline
  • Note archive: add note, design decision, analysis memo, bug fix

Intent Dispatch

User intentOperationLoad reference
"add/log [activity] [date/time]", "what's on today/this week", "mark X as done"Time-series ADD/QUERY/UPDATEdiary-interface.md
"add milestone", "mark complete", "what's at risk", "show upcoming"Status-tracked ADD/QUERY/UPDATEmulti-db-interface.md
"log experiment", "set blocker", "log findings"Experiment ADD/UPDATEmulti-db-interface.md
"add paper", "what am I reading"Reading-catalog ADD/QUERYmulti-db-interface.md
"add chapter", "move to drafting", "chapter progress"Manuscript ADD/UPDATE/QUERYmulti-db-interface.md
"add note", "design decision", "bug fix"Note ADDmulti-db-interface.md
"link paper to project", "hub and spoke", "create project view"Cross-DB relationcross-db-workflow.md
"update handoff", session endingState persistencehandoff-protocol.md

Shared Conventions

  • Title property: always use <TITLE_PROPERTY> from workspace config; never assume a fixed property name.
  • Multi-select values: pass JSON-serialized string format "[\"Value1\", \"Value2\"]", not a native array.
  • Date properties: use expanded fields date:<PROP>:start and date:<PROP>:is_datetime.
  • Default behavior: do not request confirmation for simple creates; execute and report result.
  • Find-then-update: search first with notion-search + data_source_url=collection://<DATA_SOURCE_ID>, then patch by page_id with notion-update-page.

Fallback Routing

When intent does not map cleanly to a known pattern:

  1. Read workspace configuration and enumerate available databases plus schema summaries.
  2. Match the user's noun/verb semantics to the most likely database target.
  3. Validate property names and allowed option values from that database schema.
  4. Execute the standard ADD/QUERY/UPDATE flow with runtime-resolved identifiers.

Additional Resources

  • references/diary-interface.md — Time-series log ADD/QUERY/UPDATE
  • references/multi-db-interface.md — Status-tracked, experiment, reading-catalog, manuscript, note patterns
  • references/cross-db-workflow.md — Hub-and-spoke, relation linking, filtered views
  • references/handoff-protocol.md — Session state persistence (CURRENT/NEXT/BLOCKERS)
  • references/api-patterns.md — Shared MCP API call shapes
  • references/advanced-api.md — Markdown API, pagination, rate limiting, Notion-Flavored MD, Python client patterns, advanced property shapes (salvaged from notion-io)
  • examples/schema-template.sql — Blank schema templates for common patterns

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.