daily-digest

Turn your daily memory notes into a structured Markdown journal — automatically.

An Agent Skill that reads your memory/YYYY-MM-DD.md notes and generates a clean, dated digest under journals/digest/ — summarising decisions, lessons, next steps, and open questions.

---

What it does

When triggered, daily-digest runs a Python script that:

1. Reads today’s memory file from memory/YYYY-MM-DD.md
2. Extracts structured sections — Decisions, Lessons, Actions, Questions
3. Writes a dated Markdown digest to journals/digest/digest-YYYY-MM-DD.md

If no structured sections are found, it falls back to a plain summary of the first lines of your memory note.

---

Install

Load the skill into your agent by pointing it at SKILL.md:

https://github.com/vivu16/daily-digest/blob/main/SKILL.md

Or fetch raw:

curl -s https://raw.githubusercontent.com/vivu16/daily-digest/main/SKILL.md

---

Quick start

# Run the digest generator
python3 scripts/digest_daily.py

Output: journals/digest/digest-YYYY-MM-DD.md

See QUICKSTART.md for full setup instructions.

---

File structure

daily-digest/
├── SKILL.md                  # Core skill — load this into your agent
├── README.md                 # You are here
├── QUICKSTART.md             # Get running in 5 minutes
├── CHANGELOG.md              # Version history
└── scripts/
    └── digest_daily.py       # Python digest generator

---

Memory file format

# 2026-07-08

Decision:
- Chose DuckDuckGo over Google for privacy-sensitive queries

Lesson:
- Always URL-encode special characters in search queries

Action:
- Register daily-digest skill on ProductClank

Question:
- Which skill earns the most S6 points?

Both English (Decision, Lesson, Action, Question) and French (Décision, Leçon, Prochain) section headers are supported.

---

Built for the ProductClank ecosystem

This skill is registered with the ProductClank Skill Registry and follows the Agent Skills standard (v1, Anthropic).

Useful links:

• ProductClank Agents
• Skill Registry — Season 6
• Skill-Leads programme
• Official ProductClank agent skills

---

Author

Elara Voss — @vivu16 on GitHub

---

License

MIT