why
Skill0 starsChanged 5 days agoMITView on GitHub
Use when the user asks why something was built this way or what the design rationale is. Answers in chat with sources and found, inferred, or unknown evidence; writes no tickets or code. Questions about how it works go to how.
112 other projects use a version of this.
Adopted by
Most recently taken up by
- Yeachan-Heo/oh-my-claude-sisyphustheir own version, first seen Nov 8, 2026
- Loner1024/skillstheir own version, first seen Nov 8, 2026
- caidanw/skillstheir own version, first seen Nov 8, 2026
- Yeachan-Heo/oh-my-claudecodetheir own version, first seen Nov 8, 2026
- endqwerty/endstacktheir own version, first seen Nov 8, 2026
- tristanphvn/keeltheir own version, first seen Nov 8, 2026
Used with
Projects with it also use addskill, apply-firmware and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
ryanportfolioSkilladdskill
Create, import, update, or install repo or personal skills: runtime ownership, resources, discovery checks.
Also in 6 projects that use this
- ryanportfolioSkill
apply-firmware
Bring current folder (plain folder, existing repo, old or partial firmware copy) up to full Harness Firmware in place. Mirror external repo → adopt-repo.
Also in 6 projects that use this
- +10 more from ryanportfolio
Other versions112 projects use a version of this.
Word for word · 1
- Gabriel-Lafrance/Skills★ 0100% same
Edited · 0
None found.
Wrote their own version · 111
- Yeachan-Heo/oh-my-claude-sisyphus★ 41ksame name
- Yeachan-Heo/oh-my-claudecode★ 41ksame name
- backnotprop/pstack★ 3.8ksame name
- moeru-ai/xsai★ 667same name
- jtaroreh/agystack★ 128same name
- uzairansaruzi/p3-stack★ 126same name
- Dimon94/dverity★ 107same name
- Dimon94/skills★ 107same name
- dsebban/skills★ 71same name
- Anuj7411/sipcode★ 46same name
101 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.
In this repository18 other files it gives its agents.
- AGENTS.mdAGENTS.md
- analyzeSkill
- ask-gabrielSkill
- babysit-prSkill
- councilSkill
- gabriel-modeSkill
- grill-meSkill
- hillclimbSkill
- howSkill
- review-loopSkill
- reviewSkill
- rulesSkill
- setup-gabriel-skillsSkill
- takeover-prSkill
- taskSkill
- test-auditSkill
- verificationSkill
- write-ticketSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 105 edited versions we read keep these.
3
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: line 31
Read the code that embodies it.
- Very specific: line 32
Run source control on those paths: git log, git blame, and gh for recent commits and pull requests.
- Distinctive: line 33
Only then cover the other categories below.
- Distinctive: line 69
Every search, including ones that returned nothing.
“Common” is measured: kept by at least 6 in 10 of the 105 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.
Files
Why
Answer "why is it this way?" with evidence. A fact, an inference, and a gap are different sentences.
/how answers how it works. /analyze writes the broader memo. This skill does not write tickets or code, and it does not start a build. The selected execution skill owns any later build.
The evidence tiers are adapted from poteto's pstack /why (MIT). The steps below are this pack's.
Read when
- About to search history or read large files? Open main-context.md. Skip it and raw logs bury the reason.
- About to judge whether the shape is still the right one? Open code-quality.md and code-structure.md. Skip them and the rationale endorses a messy layout.
- About to reply? Open Plain language and Unslop. Skip them and the answer sounds more certain than the evidence.
- About to ask the user anything? Open Asking the user. Skip it and you ask what git already answers.
Contract
- Anchor in the code before any parallel search: paths, symbols, then recent commits and pull requests (
git,gh). - A missing connector is a gap. Do not invent tickets, chat, metrics, or errors.
- Label every inference. Confidence matches the evidence.
Process
- Name the decision you are explaining: a shape, a name, or a rejected alternative.
- Read the code that embodies it. Record paths and symbols.
- Run source control on those paths:
git log,git blame, andghfor recent commits and pull requests. This category always runs. - Only then cover the other categories below. Parallel lookups are fine after the anchor. Use the shared roles and selection policy: explorers retrieve history; analyzers judge the evidence. Choose effort from the actual uncertainty, not the label "research".
- Standalone, write the answer below. Nested, follow nested capabilities and return the bounded decision rationale without a full standalone answer. Preserve Found / Inferred / Unknown, confidence, and source evidence in either mode. List every search you ran, including the empty ones. Keep each lookup tied to the named historical decision; do not widen into a new design investigation.
Evidence categories
Cover every row. When this session has no connector for a category, write "no connector" in Sources consulted. Do not skip the row and do not fill it.
| Category | Where to look when a tool exists |
| Source control | git and gh on the anchoring paths. Always. |
| Issue tracker | The connected tracker (GitHub issues, Linear, or whatever this session can query) |
| Long-form docs | ADRs, RFCs, and design docs in the repo or a connected doc tool |
| Team chat | The connected chat tool |
| Infra observability | Connected logs, traces, or dashboards |
| Error tracking | The connected error tracker |
| Product analytics | The connected analytics tool |
Answer
Why
The decision, in a few sentences. Each sentence is a fact or is labeled as an inference.
Evidence
- Found: paths, symbols, commit subjects, pull request numbers, issue ids. Something you opened.
- Inferred: the conclusion, and the found facts it rests on.
- Unknown: what would change the answer, and was not available.
Sources consulted
Every search, including ones that returned nothing. Name the tool and the query. A category with no connector is "no connector".
Confidence
High, medium, or low, and the one reason.
If this is a precursor to a change
Optional. Omit when the user only wanted the history.
- Preserve: what the evidence says to keep
- Change: what the evidence says is safe to change
- Avoid: the approach the history already rejected
- Risk: what the evidence does not cover
Anti-patterns
- A why with no path or symbol
- Treating an inference as a fact
- Filling a gap with a plausible story
- Starting Slack, tickets, or metrics before the code anchor
- Writing a ticket or editing code
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.