summarize-urls
Skill24 starsChanged 5 months agoMITView on GitHub
Fetch and summarize a list of URLs (e.g. from OneTab) into Obsidian-friendly markdown with a brief description of each link.
In this repository7 other files it gives its agents.
- backup-vaultSkill
- doneSkill
- extract-insightsSkill
- find-place-coordinatesSkill
- index-atomic-notesSkill
- organize-inboxSkill
- weekly-reviewSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 3 are distinctive, and 2 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.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 28
If a title is present, use it as the display name.
- Distinctive: line 35
Ask WebFetch to extract: what the page is about in 1-2 sentences, and any key takeaways.
- Very specific: line 36
If a URL is unreachable (localhost, paywalled, 404, etc.), note it as (could not fetch) and use the title if available.
- Very specific: line 53
If there are more than 5 links and clear thematic clusters emerge (e.g.
- Distinctive: line 85
Remove or re-categorize any entries
≈ “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
Summarize URLs
Purpose
Turn a batch of URLs -- typically exported from OneTab or a similar browser extension -- into a curated, Obsidian-friendly markdown summary. Each link is fetched, read, and distilled into a short description so the list becomes a useful reference note rather than a wall of raw URLs.
Instructions
1. Parse the input
The user will paste a list of URLs. Each line can be in one of these formats:
| Format | Example |
url | title | https://example.com | Example Site |
url | title | source | https://example.com | Example Site | Hacker News |
| bare URL | https://example.com |
- Split on
\|(pipe surrounded by spaces) to separate URL from title/source. - Ignore blank lines.
- If a title is present, use it as the display name; otherwise derive one from the page content.
2. Fetch and summarize each URL
For each URL:
- Fetch the page content using WebFetch.
- Ask WebFetch to extract: what the page is about in 1-2 sentences, and any key takeaways.
- If a URL is unreachable (localhost, paywalled, 404, etc.), note it as
(could not fetch)and use the title if available.
Fetch multiple URLs in parallel where possible to save time.
3. Build the markdown summary
Structure the output as:
## Link Summary
| Link | Summary |
| --- | --- |
| [Display Title](url) | 1-2 sentence summary |
| ... | ... |If there are more than 5 links and clear thematic clusters emerge (e.g. "AI / Agents", "Web Dev", "Security"), group them under subheadings instead of a single flat table:
## Link Summary
### AI / Agents
| Link | Summary |
| --- | --- |
| ... | ... |
### Web Dev
| Link | Summary |
| --- | --- |
| ... | ... |4. Append to the daily note
- Find the Obsidian vault path by running
obsidian vault. - Append the markdown summary to the daily note by running:
obsidian daily:append content="<the full markdown summary>" silent- Escape any double quotes in the content so the shell command works.
5. Confirm and offer next steps
Tell the user the summary was appended to their daily note, with a brief count (e.g. "Added 8 links to your daily note"). Then ask if they'd like to:
- Expand on any particular link
- Remove or re-categorize any entries
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.

