gborges-standard

Plugin gborges-standard from marketplace gborges in gborges0727/claude-plugins. Commands read from the repo's manifest — 2 days ago.

Skills (18)

• bear-notes

  Bear note writing. Use when saving, importing, appending, overwriting, tagging, or attaching anything in Bear, when mirroring files or docs into Bear, or when the user says to put something in their notes.

• codex-delegate

  Delegate a coding subtask to the Codex CLI on an OpenAI model (GPT-6 Luna, GPT-6 Sol, or GPT-6 Astra), spending ChatGPT-plan quota instead of Claude tokens. Use when the user's message in this session asks for Codex, names a Codex rung (consult astra, ask sol, send it to luna), or continues a Codex

• domain-modeling

  Keep the project's glossary (CONTEXT.md) and decision records (docs/adr/) sharp while a design is discussed. Use when a term is fuzzy or conflicts with the glossary, when writing or editing CONTEXT.md or CONTEXT-MAP.md, or when a decision deserves an ADR.

• grill

  Interview the user in rounds until a plan, design, or idea has no open decision left. Explicit invocation only.

• handoff

  Write the conversation up so a fresh session can continue the work, and start that session when asked. Explicit invocation only.

• investigate

  Find the cause of a hard bug or a slow path and report it without fixing it. Builds a fast check that fails on the bug first, then tests hypotheses against it. Use when the user says investigate, diagnose, or debug, or hands over a bug ticket and wants the root cause.

• model-routing-review

  Re-check which model takes which rung of the delegation ladder, on Claude Code and on Codex, after a vendor ships a model or moves a price. Rebuilds docs/model-routing.md and lists every file the ladder change touches.

• pair-debate

  Put Claude Fable 5.2 and GPT-6 Astra, both at xhigh, in a room to argue a hard problem to a solution. Explicit invocation only.

• read-aloud-prep

  Rewrite a document into spoken-English Markdown that an AI narrator reads aloud smoothly, preserving every fact. Use when the user wants a PDF, paper, Markdown file, or EPUB prepped for ElevenReader, Speechify, or any text-to-speech voice.

• research

  Send a background agent to answer a question from primary sources (official docs, source code, specs) and write the findings to a cited Markdown file in the repo. Use when the user wants a topic researched, docs or API facts checked, or reading delegated so the session can keep working.

• resolving-merge-conflicts

  Resolve an in-progress git merge or rebase conflict by reading why each side changed, keeping both intents, and finishing the merge. Use when git reports conflicts or a merge or rebase is stopped.

• routine

  Grill the user into a spec for one recurring routine (a daily check, a weekly report, a step on every new issue) that an agent could build unasked. Explicit invocation only.

• spec

  Write the conversation up as a spec file, and split it into tickets only when the work is too big for one session. Explicit invocation only.

• teach

  Teach the user a topic over many sessions, keeping the lessons, references, and what they have learned as files in the current directory. Explicit invocation only.

• wait-what

  Break the last message down in plain English, context first. Explicit invocation only.

• wayfinder

  Plan work too big for one session as a map of open questions, then answer them one session at a time until the way to build it is clear. Explicit invocation only.

• writing-for-agents

  How to structure a document an agent reads (a skill, a CLAUDE.md or AGENTS.md, a doc reached by a pointer) so the agent follows it the same way every run. Use when creating or editing a skill, a CLAUDE.md, or an AGENTS.md.

• writing-voice

  Use before shipping any artifact, whatever its length. An artifact is prose that lives on after the turn: a repo file, a PR body, a commit message, a comment, a doc, a note, a spec, a plan. Runs two passes on a draft before it ships. Chat replies never get it, since the output style alone shapes the

Slash commands (3)

• /gborges-standard:add-to-git

  Clean up scratch files, stage everything, commit, and push. Explicit invocation only.

• /gborges-standard:setup-repo

  Ask where this repo keeps agent-written documents and whether specs become GitHub issues, then write .claude/gborges-standard.json at the repo root.

• /gborges-standard:setup

  Ask which models this machine can reach and whether Claude Code may add AI attribution, write the answers to ~/.claude/gborges-standard.json and ~/.claude/settings.json, and write the Codex CLI's model and status line config.

Subagents (4)

• fable-xhigh

  Fable 5.2 at xhigh effort. Runs only when the user's own message names Fable. A hook refuses every other dispatch and points it at opus-xhigh.

• haiku-medium

  Haiku 5.5 at medium effort. Use for an edit or a run whose brief names the exact change and a command that checks it, for parallel copies of one such task across files, and for fetching a named doc page outside the codebase. Never for reading code to reach a conclusion, and never for a task that mus

• opus-medium

  Opus 5.5 at medium effort. The default for delegated work, and the floor for any task that reads code to reach a conclusion (an investigation, a diagnosis, a review, a design choice).

• opus-xhigh

  Opus 5.5 at xhigh effort. Use for a task that already failed once on opus-medium or haiku-medium, or for a task that is one long dependent chain the orchestrator cannot split into parallel pieces. Also the stand-in for fable-xhigh when the account cannot run Fable.

Similar plugins