Fetches and learns from technical blogs to improve Claude Code usage.
Install
mkdir -p .claude/skills/cc-blog && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/16763" && unzip -o skill.zip -d .claude/skills/cc-blog && rm skill.zipInstalls to .claude/skills/cc-blog
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Learn from blogs to improve Claude Code usageKey capabilities
- →read blogs from blogs.yaml
- →fetch latest articles using WebFetch
- →extract Claude Code related content
- →propose implementation steps
- →record effective content in CLAUDE.md
- →update state.yaml to mark articles as seen
How it works
The skill reads blog configurations, fetches new articles, compares them with the last check state, extracts Claude Code related content, and proposes implementation steps.
Inputs & outputs
When to use cc-blog
- →Monitor blogs
- →Learn AI patterns
- →Update internal docs
About this skill
Learn from Blogs to Improve Claude Code Usage
Read the latest articles from blogs defined in blogs.yaml and extract/practice content that can improve Claude Code usage.
Configuration
blogs.yaml- Blogs to monitor (in this skill directory)blogs.example.yaml- Configuration example
State Tracking
state.schema.json- JSON Schema definitionstate.yaml- Last check state (auto-created, gitignored)state.example.yaml- Example with schema reference
Rules:
last_checkedis Unix epoch seconds (UTC)- Schema comment (
# yaml-language-server: ...) must be preserved on updates - Concurrent execution: last write wins (use atomic write: temp file + rename)
- Orphan keys: remove entries from
state.yamlif blog is deleted fromblogs.yaml - Only update state for successfully fetched blogs
Recovery:
- If
state.yamlis corrupted, delete it and run/cc-blogagain
Tasks
- Read
blogs.yamlto get blog list - Read
state.yamlif exists (first run: treat all as new, createblogs: {})- If parse error: warn user and offer to delete and recreate
- Remove orphan keys from state (blogs not in blogs.yaml)
- For each blog:
- Fetch latest articles using WebFetch
- If fetch fails: report error and skip this blog (do not update state)
- Compare with
state.yaml.blogs.<name>:- If last_url matches: "No new articles"
- If last_url differs: mark as [NEW]
- For NEW articles: Pick up content related to Claude Code / AI usage
- Extract specific things to try in your environment
- Propose implementation steps
- Record in CLAUDE.md or docs/claude-code-routine.md if effective
- Ask user: "Update state.yaml to mark these as seen?"
- If yes, update
state.yaml(atomic write: write to temp file, then rename):- Preserve schema comment at top
- Only update successfully fetched blogs
- Set last_checked to current Unix epoch
- Set last_url, last_title, last_published from newest article
Output Format
## Articles Checked
### Status
- laiso: Last check 1738540800 (3 days ago)
- Anthropic Engineering: Last check 1738540800 (3 days ago)
### [laiso]
- [NEW] [Article Title](URL) - 2026-02-01
### [Anthropic Engineering]
- No new articles
### Errors
- (none)
## Content Related to Claude Code Usage
### Key Points
- ...
### Things to Try in Your Environment
- ...
### Implementation Steps
1. ...
2. ...
## Actions
- [ ] Try now: ...
- [ ] Add to CLAUDE.md: ...
- [ ] Consider later: ...
---
Update state.yaml to mark as seen? (y/n)
Notes
- If article is unrelated to Claude Code/AI usage, report "No related articles"
- Don't force application
- Only propose things that seem actually effective
When not to use it
- →when articles are unrelated to Claude Code or AI usage
- →when the user does not want to update state.yaml
- →when the user does not want to record content
Limitations
- →only processes blogs defined in blogs.yaml
- →only updates state for successfully fetched blogs
- →only proposes things that seem actually effective
How it compares
This skill automates the process of monitoring blogs for Claude Code patterns and suggesting actionable steps, rather than manually browsing and extracting information.
Compared to similar skills
cc-blog side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| cc-blog (this skill) | 0 | 6mo | No flags | Intermediate |
| youtube-collector | 1 | 8mo | Review | Intermediate |
| ml-paper-writing | 48 | 6mo | Review | Advanced |
| nsfc-research-content-writer | 4 | 4mo | Review | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
youtube-collector
greatSumini
유튜브 채널을 등록하고 새 컨텐츠를 수집하여 자막 기반 요약을 생성하는 skill. 사용자가 (1) 유튜브 채널 등록/관리를 요청하거나, (2) 등록된 채널의 새 영상 수집을 요청하거나, (3) 유튜브 영상 요약을 요청할 때 사용. 데이터는 .reference/ 폴더에 YAML 형식으로 저장됨.
ml-paper-writing
davila7
Write publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Use when drafting papers from research repos, structuring arguments, verifying citations, or preparing camera-ready submissions. Includes LaTeX templates, reviewer guidelines, and citation verification workflows.
nsfc-research-content-writer
huangwb8
为 NSFC 正文"(二)研究内容"写作/重构,并同步编排"特色与创新"和"三年年度研究计划",输出可直接落到 LaTeX 模板的三个 extraTex 文件。适用于用户要写/改"研究内容、研究目标、关键科学问题、技术路线、创新点、三年计划/里程碑"等场景。
update-release-notes
tldraw
Update the release notes file at `apps/docs/content/releases/next.mdx` based on PRs merged to main since the previous release, or archive `next.mdx` to a versioned file when a new version is published.
blogpost-creator
Xe
Create new blogposts from a template or linkposts using the hydrate command.
postmortem
alirezarezvani
/em -postmortem — Honest Analysis of What Went Wrong