agentleFS
orange-brotherorange-brother/obsidian-skills/skills/dev-log/SKILL.md

dev-log

Skill4 starsChanged 8 months agoMITView on GitHub

Create a retrospective Obsidian note from a development task. Given a task name and brief background, the AI collects git and code context, interviews the user about decisions and insights, and writes a structured markdown log to the vault. Use when the user wants to document a completed task, feature, or bugfix as a retrospective note.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. dimple-smile/agent-skills★ 5same name
    2. coddingtonbear/skills★ 3same 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 repository6 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 9 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    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
  1. Very specific:

    Check if VAULT_PATH is still set to the placeholder value /path/to/your/vault/Dev Logs.

    line 24
  2. Distinctive:

    Ask the user: "Obsidian vault 경로를 설정해주세요.

    line 27
  3. Very specific:

    If VAULT_PATH is already configured, skip this step silently.

    line 31
  4. Distinctive:

    If either is missing, ask for it before proceeding.

    line 39
  5. Distinctive:

    Use this context to understand the technical scope of the task.

    line 57
  6. Distinctive:

    Use the obsidian-cli skill to search for existing notes related to this task.

    line 61
  7. Distinctive:

    Use the results to identify notes worth linking to in the final output.

    line 70
  8. Distinctive:

    Ask 3–5 targeted questions to surface tacit knowledge that cannot be found in the code.

    line 74
  9. Distinctive:

    Ask all questions at once, not one by one.

    line 82
  10. Distinctive:

    Use the obsidian-markdown skill for structure — frontmatter, headings, callouts, and wikilinks.

    line 86
  11. Distinctive:

    Use the human-writer skill for all prose content within each section.

    line 88

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Dev Log Skill

Turn a completed task into a retrospective Obsidian note. The AI collects context from the repository, interviews the user, and writes a structured .md file to the vault.

Configuration

Set your vault path before using this skill:

VAULT_PATH = /path/to/your/vault/Dev Logs

Replace /path/to/your/vault/Dev Logs with the actual folder path in your Obsidian vault where dev logs should be saved.

Workflow

Step 0 — Check configuration

Check if VAULT_PATH is still set to the placeholder value /path/to/your/vault/Dev Logs.

If it is:

  1. Ask the user: "Obsidian vault 경로를 설정해주세요. dev-log 파일이 저장될 폴더 경로를 알려주세요. (예: /Users/yourname/Documents/MyVault/Dev Logs)"
  2. Once the user provides the path, update this SKILL.md file — replace the placeholder with the actual path
  3. Confirm the path is set and proceed

If VAULT_PATH is already configured, skip this step silently.

Step 1 — Receive task input

The user provides:

  • Task name: a short label for the task (used in the filename and title)
  • Background: 1–3 sentences describing what the task was about

If either is missing, ask for it before proceeding.

Step 2 — Collect git and code context

Without asking the user, run the following to gather context automatically:

git log --oneline -10
git diff HEAD~1 HEAD --stat
git diff HEAD~1 HEAD
git status

Also check for PR information if available (e.g. via gh pr view). Extract:

  • What files changed and how
  • What the commits say
  • Any PR title or description

Use this context to understand the technical scope of the task. Do not ask the user about things you can derive from git.

Step 2.5 — Find related notes in the vault

Use the obsidian-cli skill to search for existing notes related to this task. Run:

obsidian search query="{task-name}" limit=10
obsidian search query="dev-log" limit=10

This step requires Obsidian to be running. If it is not, skip silently and proceed without wikilinks.

Use the results to identify notes worth linking to in the final output. Do not ask the user about this — infer relevance from titles and content.

Step 3 — Interview the user

Ask 3–5 targeted questions to surface tacit knowledge that cannot be found in the code. Focus on:

  • Decision rationale: Why this approach over alternatives?
  • Tradeoffs: What did you give up or accept?
  • Surprises: What was harder or easier than expected?
  • Insights: What would you do differently next time?
  • Open ends: What's left unresolved or could follow from this?

Ask all questions at once, not one by one. Keep questions short and concrete based on what you learned from the git context.

Step 4 — Write the dev log

Use the obsidian-markdown skill for structure — frontmatter, headings, callouts, and wikilinks.

Use the human-writer skill for all prose content within each section. Apply its vocabulary, structural, and voice rules to every sentence written. The text should sound like the user wrote it, not like an AI summarized it.

Frontmatter — include title, date, tags, status, project, and repo fields:

---
title: {Task Name}
date: {YYYY-MM-DD}
tags:
  - dev-log
project: {project or repo name}
repo: {org/repo}
status: done
---

Wikilinks — link related notes, projects, or previous dev-logs using [[Note Name]] wherever relevant. If the task relates to a known project or previous task, link to it.

Callouts — use callouts to highlight key moments in the log:

  • > [!insight] for key learnings and takeaways
  • > [!question] for unresolved questions or open ends
  • > [!warning] for tradeoffs or things to watch out for

Save the file to:

{VAULT_PATH}/{YYYY-MM-DD}-{task-name}.md

Where {task-name} is the task name lowercased with spaces replaced by hyphens.

Output Structure

frontmatter   → title, date, tags: [dev-log], project, repo, status: done
# {Task Name}
## 배경            → user's description, lightly expanded
## 변경 내역        → bullet list from git context, with [[wikilinks]] where relevant
## 고민과 판단      → key decisions from interview, written in first person
               → use > [!warning] for tradeoffs
## 인사이트         → use > [!insight] for each key learning
## 다음 액션        → use > [!question] for unresolved items, omit if nothing

You MUST write the entire log in Korean, regardless of the language the user used. Keep each section concise — this is a log, not documentation.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.