Imported from ahkohd/dotfiles (home/AGENTS.md). Install upstream with npx skills add ahkohd/dotfiles --skill home. Copyright stays with the author.

Guidelines

Tool Use

• Use gh CLI for GitHub operations: viewing PRs, fetching diffs, checking issues, browsing repo contents, API calls.
• Use fd and rg for filesystem and text search. Prefer fd over find, and rg over grep.

VCS

• Use jj for version control. Prefer jj over git where possible.
• Disable GPG signing when running jj write commands. Always pass --config 'signing.behavior="drop"' to commands like jj commit, jj describe, jj new, jj bookmark, etc.
• Disable GPG signing for git commits. Always pass -c commit.gpgsign=false to git write commands. E.g. git -c commit.gpgsign=false commit -m "message".

Writing Style

• Never use + or & as conjunctions in prose, commit messages, or comments. Write "and" instead. E.g. "add debug and tracking", not "add debug + tracking".
• For reports, summaries, guidance, or other prose, use the govuk-style skill.
• Stick to plain ASCII in text. No unicode symbols that aren't on a standard keyboard. E.g. write "to" not "→", write "-" not "•", write ">=" not "≥".
• Be concise. Humans skim — cut filler, keep it short. This includes code comments, documentation, commit messages, PR descriptions, and chat responses.
• No emojis. Not in commit messages, comments, or docs.

Picking the right models for workflows and subagents

Rankings, higher = better. Cost reflects what I actually pay, not list price. Intelligence is how hard a problem you can hand the model unsupervised. Taste covers UI/UX, code quality, API design, and copy.

How to apply:

• These are defaults, not limits. You have standing permission to override them: if a cheaper model's output does not meet the bar, rerun or redo the work with a smarter model without asking. Judge the output, not the price tag. Escalating costs less than shipping mediocre work.
• Cost is a tie-breaker only; when axes conflict for anything that ships, intelligence > taste > cost.
• Bulk/mechanical work (clear-spec implementation, data analysis, migrations): gpt-5.6-sol.
• When using gpt-5.6-sol & gpt-6-astra, set Pi thinking level to medium.
• When using fable-5.1, set Claude CLI thinking level to medium.
• Anything user-facing (UI, copy, API design) needs taste >= 7.
• Reviews of plans/implementations: fable-5.1, optionally gpt-6-astra as an extra independent perspective.
• Mechanics: gpt-* is handled natively by Pi through the OpenAI Codex provider, automatically adopting user-level Pi configuration from ~/.pi/agent/settings.json. Avoid custom bash wrappers. Use Pi model selection and herdr skills:
  • /model - Change the current session model.
  • /skill:model-selection - Choose model, provider and harness.
  • /skill:herdr-subagents - Spawn async subagents in background herdr workspaces. Defaults to Pi.
  • /skill:herdr-messaging - Message or reply to existing agents.
  • /skill:herdr-handoff - Create a fresh human-facing herdr session when context needs to transfer.
• For Anthropic models (sonnet-5, opus-5, fable-5.1), prefer Claude CLI.

Subagents:

• Use herdr-subagents by default for async grunt work, review, search, testing and investigations.
• For closed-loop quality assurance, spawn a fresh reviewer with herdr-subagents before finalizing.