Agent skill
compose-digest
Compose a formatted news digest from analyzed articles and trends. Uses memory search to retrieve articles and LLM to generate formatted output.
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/compose-digest
Metadata
Additional technical details for this skill
- domain
- news
- category
- publishing
- confidence
- 0.95
- mcp servers
-
[]
- requires approval
- NO
SKILL.md
Compose Digest
Compose a formatted news digest from analyzed articles and trends.
When to Use
Use this skill when you need to:
- Create daily or weekly news digests
- Format articles for publication
- Group articles by category or topic
- Highlight trending topics
- Generate human-readable summaries
Instructions
Step 1: Retrieve Articles from Memory
Use memory/search to get recent analyzed articles:
- namespace: "news/analyses"
- query: "AI news articles" (or date-specific query)
- limit: 50 (adjust for daily/weekly)
Filter by date range if specified.
Step 2: Retrieve Trends from Memory
Use memory/search to get current trends:
- namespace: "news/trends"
- query: "trending topics"
- limit: 20
Step 3: Group Articles by Category
Organize articles into sections:
- security - Security vulnerabilities and updates
- product - New releases and product updates
- research - Latest research papers and breakthroughs
- business - Company news and market developments
- policy - Regulations and policy changes
- general - Other AI news and updates
Sort articles within each category by importance_score (highest first).
Step 4: Format with LLM
Use use_llm to compose the final digest:
Digest composition prompt:
Compose a news digest from these articles and trends.
ARTICLES BY CATEGORY:
{categorized_articles}
TRENDING TOPICS:
{trends}
Format as Markdown with:
1. Header with title and date
2. Breaking news section (importance >= 9)
3. Trending topics section
4. Category sections in order: security, product, research, business, policy, general
For each article include:
- Title with importance indicator (🔴 critical, 🟠 important, 🟡 notable, ⚪ minor)
- Source name
- 2-3 sentence summary
- Link to source
Keep summaries concise. Use emojis for section headers.
Step 5: Return Formatted Digest
Return the composed digest in Markdown format.
Tool Usage Guidance
memory/search
- Search "news/analyses" for recent articles
- Search "news/trends" for trending topics
- Filter by date range as needed
use_llm
- Use for final formatting and composition
- Provide all articles and trends as context
- Request structured Markdown output
Digest Structure
# AI News Digest
**Date:** 2026-01-31
**Articles:** 25 | **Trends:** 8
---
## 🚨 Breaking News
### GPT-5 Released with AGI Capabilities
*Major model release from leading AI company*
OpenAI has released GPT-5, claiming significant advances...
[Read more](https://...)
---
## 🔥 Trending Topics
### Breaking
- **GPT-5** (15 mentions)
- **AGI** (12 mentions)
### Hot
- **Claude 4** (8 mentions)
- **Gemini Pro** (7 mentions)
---
## 🔒 Security
### 1. 🔴 Critical Vulnerability in Popular AI Framework
**Source:** Security Research Blog
Researchers have discovered a critical vulnerability...
[Read more](https://...)
---
Category Order (Priority)
- Security - Most urgent (vulnerabilities, incidents)
- Product - Time-sensitive (launches, releases)
- Research - Valuable but not urgent
- Business - Market context
- Policy - Regulatory updates
- General - Everything else
Importance Indicators
| Score | Indicator | Meaning |
|---|---|---|
| 9-10 | 🔴 | Critical - Major news |
| 7-8 | 🟠 | Important - Significant impact |
| 5-6 | 🟡 | Notable - Meaningful development |
| 1-4 | ⚪ | Minor - Incremental update |
Length Guidelines
| Digest Type | Articles | Pages |
|---|---|---|
| Daily | 10-20 | 2-3 |
| Weekly | 30-50 | 5-10 |
| Breaking | 1-3 | 1 |
Error Handling
- If search returns no articles, return empty digest with explanation
- If LLM formatting fails, return raw article list
- Log but continue if individual articles have issues
Success Criteria
- Digest is well-structured and readable
- Articles grouped logically by category
- Important stories prominently featured
- Trending topics highlighted
- Length reasonable for consumption
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?