What does this skill do, and when should you use it?
This skill researches across multiple Notion sources and turns the findings into structured documentation. It searches for relevant pages, fetches their contents, and records key facts, citations, and source links. It supports quick briefs, summaries, comparisons, and comprehensive reports. The resulting document can be created in Notion and updated as new information arrives.
Searches Notion for targeted sources, fetches relevant pages, captures facts, metrics, claims, constraints, and dates, tracks source URLs or IDs, selects an output format and matching template, creates a Notion page with inline citations and a references section, and updates pages with new findings.
- A team member needs project background consolidated from several Notion pages.
- An analyst needs scattered internal material synthesized into a research brief.
- A product or operations lead needs an option comparison published in Notion with source citations.
- A team needs to maintain an evolving research document containing highlights, risks, and open questions.
How do you install this skill?
- Core functionality depends on a remote Notion MCP; the supplied material does not establish reliable reachability from mainland-China networks.
- The skill can create and update pages and may create tasks; explicit per-action confirmation, permission scoping, sensitive-data handling, and rollback should be added or verified before use.
- The reference files cited by SKILL.md are not included in the supplied material; verify that those files, tool interfaces, and templates actually exist after installation.
- External research, pricing, market, and travel claims in the examples lack verifiable sources and should not be treated as evidence that the skill performs external fact checking.
- Network access
- Local filesystem
- MCP Server
Notion MCP
The skill is located at skills/.curated/notion-research-documentation/. The README says curated skills can be installed in Codex with $skill-installer notion-research-documentation; restart Codex after installation. The source does not document the exact destination path beyond the installer behavior.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/openai/skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/.curated/notion-research-documentation" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Research the Notion pages about project X, produce a cited comparison report, and publish it to Notion.
A suitable trigger could be: “Research the Notion pages about project X, produce a cited comparison report, and publish it to Notion.” The workflow searches Notion first, then fetches pages; when multiple results appear, it requires scope confirmation. The referenced reference/ directory contains the detailed format and template guidance.
What are this skill's strengths and limitations?
- Defines an end-to-end workflow from source discovery through Notion publication.
- Explicitly preserves source links, inline citations, and a references section.
- Supports quick briefs, summaries, comparisons, and comprehensive reports.
- Supports ongoing maintenance through Notion page updates.
- Requires Notion MCP; the source instructs the user to configure it and complete OAuth if calls fail.
- The skill documents Notion research and documentation, but does not establish support for external websites, databases, or other knowledge bases.
- The supplied source includes no test suite, platform validation record, or individual LICENSE.txt contents.
- The parent openai/skills repository is marked deprecated in its README.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Notion Research & Documentation this page ✓ OpenAI · Official | 43 · Not recommended | ★ 28k | 4mo ago | — |
| Notion Knowledge Capture ✓ OpenAI · Official | 36 · Not recommended | ★ 28k | 4mo ago | — |
| Notion Spec-to-Delivery Planner ✓ OpenAI · Official | 43 · Not recommended | ★ 28k | 4mo ago | — |
| Notion Meeting Intelligence ✓ OpenAI · Official | 38 · Not recommended | ★ 28k | 4mo ago | — |
| Obsidian Second Brain | 40 · Not recommended | ★ 4.8k | 15d ago | MIT |
How did FollowSkills review this skill?
The workflow requires searching, fetching, and creating or updating Notion pages, and it discloses the MCP and OAuth data path. However, it only clearly requests confirmation of research scope; per-write confirmation for publishing, task creation, or updates is incomplete, and sensitive-data handling, least privilege, rollback, and recovery are unspecified, so points are deducted.
The main workflow is reasonably consistent and includes a response to an unconnected Notion MCP. However, referenced files are not present in the supplied material, tool parameters and template reproduction are incomplete, and diagnostics or recovery for malformed input, conflicting sources, permission failures, and write failures are thin. Static calibration limits this score to 10 or below.
The audience, Notion multi-source research use case, report types, and scenarios are fairly clear. Non-fit boundaries, precise trigger conditions, Chinese-language output support, and mainland-China network reachability are not established; the core capability depends entirely on Notion MCP, so points are deducted.
The skill provides a quick start, staged workflow, examples, and evaluation guidance with useful information architecture. However, referenced materials are missing, and versioning, changelog, license, maintenance ownership, and update path are not clear; the supplied metadata also says the individual skill license is unknown, so points are deducted.
The claimed outputs—structured briefs, comparisons, reports, and citations—are clear, and the examples show directly usable document structures. Yet example correctness and tool execution are not statically verified, while database properties, templates, and citation formatting may require manual correction. Static calibration limits this score to 7 or below.
Evaluation scenarios, success criteria, and examples provide limited auditable evidence. There is no real CI, committed execution result, or independent corroboration; the evaluation files mostly specify expected behavior rather than verified outcomes. Static calibration limits this score to 5 or below.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →