ai-digest
Skill1 starsChanged 2 months agoMITView on GitHub
Generates a weekly AI news and research digest. Researches the latest in AI models, tools, research papers, and products from curated sources in config.digest.sources, then produces a structured HTML digest with a personal editor's note.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- caoergou/erics-skillstheir own version, first seen Apr 10, 2026
- airborne12/agent-skillstheir own version, first seen Mar 14, 2026
- alxyrgin/secondbrain_aitheir own version, first seen Feb 18, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- alxyrgin/secondbrain_ai★ 17same name
- airborne12/agent-skills★ 6same name
- caoergou/erics-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository10 other files it gives its agents.
- .agents/AGENTS.mdAGENTS.md
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- daily-digestSkill
- humanizeSkill
- linkedinSkill
- meSkill
- newsletterSkill
- pixelSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. About 0 are common, 5 are distinctive, and 8 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
8
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 26
Create this checklist first.
- Very specific: line 40
Load your profile & memory: Read config.profile (name, role, voice, brand, content pillars), the voice samples in content.md when tone …
- Distinctive: line 41
Ask if anything is missing: If a required input is unknown, stop and ask a specific question.
- Very specific: line 43
Validate, output & record: Check the result against this skill's rules, return it, then append any new durable fact to memory.md and any …
- Distinctive: line 47
Before generating, optionally pre-fetch source content:
- Distinctive: line 54
If you skip this step, use WebFetch/WebSearch to gather content directly.
- Very specific: line 65
Read sources list: Read config.digest.sources — curated AI sources with categories, URLs, RSS feeds.
- Very specific: line 66
Read author profile: Read config.profile — for editor name, title, and ai_digest_prefs.
- Very specific: line 67
Determine the week: Use $ARGUMENTS as the date (YYYY-MM-DD) if provided, otherwise use today.
- Very specific: line 68
Check for pre-fetched data: Look for output/ai-digest/fetched-{date}.json — use it if present.
- Distinctive: line 89
Use pre-fetched data if available, otherwise WebFetch/WebSearch
- Distinctive: line 129
Use clean, readable HTML:
- Very specific: line 138
Write the complete HTML file to output/ai-digest/digest-{date}.html
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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
AI Weekly Digest Skill
Role
You are an AI-news curator. You research the week's AI developments and assemble a self-contained HTML digest with a personal editor's note.
Context to load
Load and honor these before acting:
config.profile— identity, role, voice, brand, content pillars (personalize everything)content.md— voice samples and prompt guidance (when tone matters)memory.md/notes.md— learned preferences and open follow-upsconfig.digest.sources— curated sources and categories
Workflow
Step 0: TodoWrite Checklist
Create this checklist first; mark one task in_progress, finish it, then move to the next.
TodoWrite([
{ content: "Gather inputs; load config.profile, content.md, memory.md & notes.md", status: "in_progress" },
{ content: "Ask for any missing inputs — never assume or hallucinate", status: "pending" },
{ content: "Research the sources and build the weekly digest HTML", status: "pending" },
{ content: "Validate output; record new facts to memory.md / follow-ups to notes.md", status: "pending" }
])Steps
- Gather inputs. Parse the request /
$ARGUMENTSand list what this skill needs. - Load your profile & memory. Read
config.profile(name, role, voice, brand, content pillars), the voice samples incontent.mdwhen tone matters, and the sharedmemory.md/notes.mdfor learned preferences and open follow-ups. - Ask if anything is missing. If a required input is unknown, stop and ask a specific question. Never assume, guess, or invent facts, numbers, names, or sources.
- Do the work following the sections below.
- Validate, output & record. Check the result against this skill's rules, return it, then append any new durable fact to
memory.mdand any follow-up tonotes.mdviapython .agents/scripts/memory.py.
Prerequisites
Before generating, optionally pre-fetch source content:
python .agents/scripts/ai-digest/fetch_sources.py [--date YYYY-MM-DD]This fetches RSS feeds and web pages into output/ai-digest/fetched-YYYY-MM-DD.json.
If you skip this step, use WebFetch/WebSearch to gather content directly.
Automation (run every Monday)
python .agents/scripts/ai-digest/run_digest.py --schedule
# Then: python run_digest.py [--date YYYY-MM-DD] [--open]Setup — always do this first
- Read sources list: Read
config.digest.sources— curated AI sources with categories, URLs, RSS feeds. - Read author profile: Read
config.profile— for editor name, title, andai_digest_prefs. - Determine the week: Use
$ARGUMENTSas the date (YYYY-MM-DD) if provided, otherwise use today. - Check for pre-fetched data: Look for
output/ai-digest/fetched-{date}.json— use it if present.
Digest categories
Research and gather content for each category (matching digest_categories in sources.json):
| Category | What to cover |
breaking_news | Major AI announcements, funding rounds, launches |
models_apis | New model releases, API updates, benchmark results |
research_papers | Notable papers from arXiv, Anthropic, OpenAI, Google DeepMind |
github_repos | Trending AI repos, new open-source tools |
products_tools | New AI products, apps, developer tools |
industry_analysis | Market trends, enterprise adoption, regulation |
agents_automation | Agent frameworks, agentic workflows, automation |
tutorials_learning | Notable tutorials, courses, guides |
linkedin_insights | Thoughtful takes from AI practitioners |
Research process
For each source in config.digest.sources where linkedin_worthy: true:
- Use pre-fetched data if available, otherwise WebFetch/WebSearch
- Prioritise sources by
digest_categoriesrelevance - Aim for 2–5 items per major category
Digest structure
Generate a complete HTML file at output/ai-digest/digest-YYYY-MM-DD.html.
1. Header
- Digest title, week date range, author name and title (from profile)
2. Editor's note (personalised intro)
- 80–120 words, the author's personal take on the week's biggest theme
- Conversational and opinionated — use
writingStyletone from profile - Reference 1–2 specific stories
3. Top stories (breaking_news + models_apis)
- 3–5 items. Each: headline, 2–3 sentence summary, source link, why it matters
4. Research highlights (research_papers)
- 2–3 notable papers. Each: title, authors, one-paragraph plain-English summary
5. Tools & repos (github_repos + products_tools)
- 4–6 items. Each: name, what it does, link, who it's for
6. Agents & automation (agents_automation)
- 2–4 items showing trends in agentic AI
7. Industry pulse (industry_analysis)
- 2–3 business/market takes
8. Worth reading (tutorials_learning + linkedin_insights)
- 3–5 curated links with one-line descriptions
9. LinkedIn post pick
- Select the single most "LinkedIn-worthy" story (
linkedin_worthy: true) - Draft a 150-word LinkedIn post about it in the author's voice
HTML design
Use clean, readable HTML:
- System font stack, max-width 800px, good line height
- Section headings, card-style items
- Source attribution links on every item
- Dark-mode support (
prefers-color-scheme: dark) - Author branding from profile (name, title) in footer
Output
- Write the complete HTML file to
output/ai-digest/digest-{date}.html - Report: date range, item count per category, file path
- Pull out the LinkedIn post pick for quick access
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.

