logmind
Skill1 starsChanged 4 months agoMITView on GitHub
MUST be loaded for any task in a project that uses logmind (detect by: .logmind/config.yml at repo root, or AGENTS.md / CLAUDE.md mentioning logmind, or docs/decisions.md present). Use BEFORE writing >20 lines of new code, BEFORE choosing between alternatives, BEFORE adding a dependency, BEFORE modifying existing functionality, BEFORE making any security or performance trade-off, BEFORE renaming or moving any significant module. Logging is part of the work, not after it. Also use to read prior decisions before starting any task in such a project so you don't re-litigate something already decided.
- Pipes a download into a shell
- Installs packages
- Commits and pushes
1 other project uses a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- thrillmade/agent-skillsedited, first seen Nov 8, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 1
- thrillmade/agent-skills★ 316% same
Wrote their own version · 0
None found.
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.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- clud-bug-collaborationSkill
- critical-issues-onlySkill
- evidence-based-reviewSkill
- respect-existing-conventionsSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. About 0 are common, 7 are distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 17
If the current project contains .logmind/config.yml or its AGENTS.md mentions logmind, this skill applies.
- Distinctive: line 27
Write significant new code (> 20 lines or a new module)
- Distinctive: line 29
Add or remove a dependency
- Distinctive: line 32
When in doubt, log it. A short decision entry is cheap.
- Very specific: line 52
When you are on a feature branch (anything other than the project's default branch), the entry is written to …
- Distinctive: line 61
Before starting non-trivial work, read in order:
- Very specific: line 100
Add [skip-logmind] to the commit subject.
- Very specific: line 123
After a successful commit, logmind log may print advisories to stderr — a stale component (workflow/hook/AGENTS.md drift.
- Very specific: line 131
If the project doesn't yet have logmind:
- Distinctive: line 145
Don't log every tiny edit.
- Distinctive: line 146
Don't write the decision after the fact in past tense for trivial code.
- Distinctive: line 147
Don't reword a decision someone else already logged — link or extend it.
- Very specific: line 148
Don't bypass the auto-commit (--no-commit) unless you know the project's branch protection requires it.
≈ “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.
Files
logmind: log decisions as you work
If the current project contains .logmind/config.yml or its AGENTS.md
mentions logmind, this skill applies. Log a decision before writing
non-trivial code, not after.
When to log
Log a decision whenever you:
- Make an architectural or design choice
- Choose between alternative approaches
- Write significant new code (> 20 lines or a new module)
- Modify existing functionality in a non-obvious way
- Add or remove a dependency
- Make a security or performance decision
When in doubt, log it. A short decision entry is cheap; recovering missing context months later is not.
How to log
logmind log "Use PostgreSQL for primary database" \
-r "Need ACID compliance and complex joins" \
-a "MongoDB" -a "SQLite" \
-i "Connection pooling required" \
-i "Schema migrations needed"logmind log is the commit primitive: default --stage all sweeps the
whole working tree into one commit (the decision and the code that
prompted it, together). Pass --stage scoped to commit only the decision
file + its companion docs when you have unrelated WIP.
Branch-aware logging
When you are on a feature branch (anything other than the project's default
branch), the entry is written to
docs/decisions-branches/<sanitized-branch>.md rather than
docs/decisions.md. On PR merge, a GitHub Action appends a one-line
summary linking the PR + the per-branch file to docs/decisions.md. You do
not need to manage this routing — logmind log does it automatically.
Reading prior context
Before starting non-trivial work, read in order:
docs/timeline.md— the main-canonical, source-derived union of every decision across every branch; start here.docs/decisions.md— direct-on-main decisions in detail (20 most recent).docs/decisions-branches/<your-branch>.md(if present) — decisions made earlier on the same feature branch.docs/file-structure.md— current project tree.- The project's spec file, if configured (
context.spec_filein.logmind/config.yml, typicallydocs/spec.md) — the forward-looking contract to build toward, not just the history behind you.
logmind context bundles the file structure + timeline (and the spec
file, when configured) into one cache-friendly read — prefer it over
piecing this together by hand.
logmind show # recent decisions on the current branch
logmind show --all # include archive
logmind search "postgres" # full-text across recent + archiveAs an agent, set LOGMIND_QUIET=1 for terse, chainable machine output on
the read/emit verbs (doctor, file-structure, guard-commit,
headline, log, repomap, search, show, timeline): each
suppresses progress chatter and prints a single ok <key=value> line.
Other verbs currently ignore this flag.
Enforcement: raw git commit is blocked
A substantive commit that bypasses logmind log is blocked, not just
discouraged. Two layers enforce it: the git commit-msg hook, and (inside
Claude Code) a PreToolUse hook that intercepts the git commit Bash call
before it runs. Use logmind log in place of git add + git commit +
git push.
Escape hatches, for genuinely no-decision commits (typo fixes, dependency bumps):
- Add
[skip-logmind]to the commit subject. - Set
LOGMIND_ALLOW_GIT_COMMIT=1for one command. git.enforce_commits: falsein.logmind/config.ymldisables enforcement for the whole repo.
Branch summaries (headline)
Set a one-sentence, plain-English summary of what the whole branch does — the canonical timeline shows this line for the branch, and it's the first thing the next agent reads:
logmind headline "Add JWT session auth with refresh-token rotation"
# or bundle it into a decision commit:
logmind log "Wire refresh-token rotation" -r "..." -H "Add JWT session auth with refresh-token rotation"A no-op on the default branch. logmind doctor --fix backfills a headline
for any branch file that's missing one.
The pulse: read the advisories after logmind log
After a successful commit, logmind log may print advisories to stderr —
a stale component (workflow/hook/AGENTS.md drift; run logmind doctor
--fix) or spec staleness (the spec file hasn't been touched in 20+
decisions; worth a review). These never block the commit that already
landed — act on them anyway.
Setup (one-time, per project)
If the project doesn't yet have logmind:
brew install thrillmade/tap/logmind # or: curl -fsSL https://logmind.dev/install.sh | bash
logmind init # scaffolds docs/, AGENTS.md, GH Actions, hooks
logmind doctor # confirm a clean installDon'ts
- Don't run
git add+git commitdirectly for changes that carry a decision — the enforcement hooks above will block it anyway.logmind logwrites the decision file, stages the working tree, and creates the commit in one step. - Don't log every tiny edit. The 20-line rule is a guideline; use judgement.
- Don't write the decision after the fact in past tense for trivial code.
- Don't reword a decision someone else already logged — link or extend it.
- Don't bypass the auto-commit (
--no-commit) unless you know the project's branch protection requires it.
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
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.

