swastijswastij/claude-log-skill/skills/log/SKILL.md

log

Skill16 starsChanged 23 days agoMITView on GitHub

Log this session's learnings (decisions, insights, questions, quiz) for a task to the personal Notion "Engineering learnings" databases (any task tracker). Use when the user says /log, "log this", "save learnings", or when a PR is opened or a task reaches Done and the user wants the insight captured.

18 other projects use a version of this.

Adopted by

Most recently taken up by

mehboobali98/rmine-skillstheir own version, first seen Nov 6, 2026
myowinthein/claude-helmtheir own version, first seen Nov 1, 2026
tjmode/koovitheir own version, first seen Nov 1, 2026
rmarji/chiron-counciltheir own version, first seen Nov 1, 2026
foxrun-ai/foxruntheir own version, first seen Nov 1, 2026
gyroid-eth/orrery-telemetrytheir own version, first seen Oct 27, 2026

Other versions18 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 18
regent-vcs/re_gent★ 789same name
dtiger1889-ops/claude-harness-toolbox★ 31same name
gyroid-eth/orrery-telemetry★ 12same name
foxrun-ai/foxrun★ 6same name
bonez-io/re_gent★ 4same name
Valmor-R-Reichel/SGM★ 4same name
aniklavida/agent-assemble★ 1same name
llewellynvz/claude-code-timestamps★ 1same name
rmarji/chiron-council★ 1same name
thensls/nsls-personal-toolkit★ 1same name
8 more not listed
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.

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

0
Common
Most of the 16 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Distinctive:
Don't put anything secret (tokens, customer data) in an entry.
line 31
Very specific:
Don't set status unless the user confirms it.
line 32
Distinctive:
Don't invent decisions the session didn't make.
line 33

“Common” is measured: kept by at least 6 in 10 of the 16 edited versions of this file we could read. “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

/log — capture learnings to Notion

Personal knowledge base, not team-facing. Script: ~/.claude/scripts/notion-log.py. Entry files go in the scratchpad dir (or a temp dir), named log-<task_id>.json. Setup and the Notion template: https://github.com/swastij/claude-log-skill

Steps

Identify the task. Task ID (ClickUp, Jira, Linear, GitHub issue; any id works, it is only the upsert key) from the branch name, PR body or the user. If none exists, ask. Collect PR URLs for every repo touched.
Draft the entry from the session. Write it as JSON (shape in the script docstring) and SHOW it to the user as a short readable summary, not raw JSON. Rules:
- surprised_me / how_it_works / would_bite_again: one paragraph each, mechanism not diff. Leave empty if the task was trivial; never pad.
- decisions: every product/architecture/security choice made this session, each with why and alternatives rejected. Workarounds and deferred items are decisions too (kind: workaround|deferred).
- questions with source: "I asked": notable questions the user asked that changed the outcome. Rate 1–5 by how much it mattered.
- questions with source: "should have asked": questions that, had they been asked earlier, would have saved rework or caught a bug. Rate 1–5 by cost of not asking. Be honest; this is the most valuable list.
- spawned_from: set when this task was opened because of another logged task.
- risk: high for auth, payments, migrations, infra, data deletion.
If status is Done (merged / shipped): add 3–5 source: "quiz" questions with correct_answer filled and my_answer empty. Make them about mechanism and failure modes ("what happens if X", "why not Y"), not trivia. Set shipped_on to today.
Wait for the user's edits. Apply them to the JSON. Do not send before they confirm.
Send: python3 -I ~/.claude/scripts/notion-log.py <file>. Report the script's output lines verbatim. Re-running is safe: task is upserted by Task ID, rows with an existing title are skipped.

Don'ts

Don't put anything secret (tokens, customer data) in an entry.
Don't set status unless the user confirms it; the task tracker is the source of truth.
Don't invent decisions the session didn't make.

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.

Check your setup, free