grug-review
Skill1 starsChanged 4 months agoMITView on GitHub
Delete-first review of a branch, commit, staged change, working diff, or repo. Treats lines as carrying cost; applies YAGNI, KISS, and Ousterhout-style depth checks. Use when the user asks for a grug, minimal, YAGNI, or liability review.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- davojta/skillstheir own version, first seen Nov 1, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- davojta/skills★ 1same 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.
- AGENTS.mdAGENTS.md
- bktSkill
- branch-review-canvasSkill
- pairSkill
- terraformSkill
- walkthroughSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 16
Prefer shrinking the scoped change over drive-by refactor.
- Distinctive: line 19
Prefer narrow interfaces that hide complexity instead of leaking it through signatures, shapes, or call patterns.
- Distinctive: line 20
Keep names meaningful; reserve comments for non-obvious invariants.
- Distinctive: line 21
Treat dense special cases as evidence of a mistaken model.
- Distinctive: line 33
If rollout, coupling, semantics, or ownership survive exploration unanswered, ask one concrete trade-off question at a time and include the …
≈ “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
Grug review
Subtract until correctness or a stated requirement prevents it. Reshaping inside that guardrail must not gratuitously widen surfaces.
Scrutiny:
- Prefer shrinking the scoped change over drive-by refactor.
- Flag speculative genericity, unrequested toggles, needless policy tax, dormant paths, tighter coupling, or wider blast radius.
- Duplicate before wrong abstraction; reject ceremonial structure.
- Prefer narrow interfaces that hide complexity instead of leaking it through signatures, shapes, or call patterns.
- Keep names meaningful; reserve comments for non-obvious invariants.
- Treat dense special cases as evidence of a mistaken model.
Get the diff via attached context when present; otherwise git (status, diff, logs, merge-base). One clarifying prompt if comparison base unclear—never assume silently. Multiple roots: rinse per checkout; harp on duplication or drifting contracts.
Fan out narrow parallel explore-style probes when mental model/call graphs/conventions/outside hunks justify it—not one sprawling agent—and stay read-only unless the user commissioned edits.
Order
Semantics and safety first (bad cuts void the mandate to delete); decide what truly must exist; then cross-cutting leaks and misplaced complexity; cosmetic churn only when it hides risk.
Continuous question: fewer types/files/parameters/states/layers?
If rollout, coupling, semantics, or ownership survive exploration unanswered, ask one concrete trade-off question at a time and include the recommended answer. Do not ask about discoverable facts such as call sites, prior patterns, or test coverage.
Output stays tight—cite touched files/lines. One-paragraph verdict; subtraction bullets naming what disappears or slips to a later PR; seam notes where depth/hiding/naming/special cases miss the lenses (only actionable observations); unexplored-unknowns condensed to minimal follow-up questions—no fluff praise unless liability clearly dropped.
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.