Name: .dotfiles
Author: aaronlord

Imported from aaronlord/.dotfiles (AGENTS.md). Install upstream with npx skills add aaronlord/.dotfiles. Copyright stays with the author.

Investigating test failures

When running test suites to investigate failures, run the command bare (no shell redirection), capture output from the bash result, then write it to a temp file (e.g. /tmp/test-results.txt) using the write tool. Then read or grep that file. Never run the suite multiple times with different grep patterns against live output.

Bash command habits

Do not prepend cd <project-dir> && to bash commands — the shell is already in the correct working directory. Run one command per bash tool call. Do not chain commands with &&. Wait for the result before running the next command.

Directory traversal

Use tree or find to explore directory structures instead of chaining multiple ls commands.

Code comments

Keep comments terse — caveman-style, few words, no restating the obvious. One line for guard/edge-case comments, not multi-sentence prose. State only the non-obvious why.

{ "ocm": "1", "id": "aaronlord-dotfiles-dotfiles", "kind": "skill", "name": ".dotfiles", "description": "When running test suites to investigate failures, run the command bare (no shell redirection), capture output from the bash result, then write it to a temp file (e.g. `/tmp/test-results.txt`) using the write tool. Then read or grep that file. Never run the suite multiple times with different grep patterns against live output.", "publisher": "aaronlord", "version": "1.0.0", "capabilities": { "domains": [ "general" ], "tags": [ "skill-md", "github-agents" ], "languages": [ "en" ] }, "quality_prior": 0.6, "examples": [ "When running test suites to investigate failures, run the command bare (no shell redirection), capture output from the bash result, then write it to a temp file (e.g. `/tmp/test-results.txt`) using the write tool. Then read or grep that file. Never run the suite multiple times with different grep patterns against live output." ], "primary": false, "metadata": { "source": { "provider": "github-agents", "repository": "https://github.com/aaronlord/.dotfiles", "path": "AGENTS.md", "ref": "d70b2e9562953347b56c81eca399d88a4733a5db", "url": "https://github.com/aaronlord/.dotfiles/blob/d70b2e9562953347b56c81eca399d88a4733a5db/AGENTS.md", "key": "aaronlord/.dotfiles/AGENTS.md" } }, "instructions": "### Investigating test failures\n\nWhen running test suites to investigate failures, run the command bare (no shell redirection), capture output from the bash result, then write it to a temp file (e.g. `/tmp/test-results.txt`) using the write tool. Then read or grep that file. Never run the suite multiple times with different grep patterns against live output.\n\n### Bash command habits\n\nDo not prepend `cd <project-dir> &&` to bash commands — the shell is already in the correct working directory.\nRun one command per bash tool call. Do not chain commands with `&&`. Wait for the result before runnin", "cost": { "context_tokens": 233 } }

.dotfiles

About

Investigating test failures

Bash command habits

Directory traversal

Code comments

Use it

Manifest

Reviews