knowledge-base

Some AI Stuff that Could Be Useful

Warp — An agentic development environment combining a modern terminal, code editor, and AI coding agents.
CodeBurn — A local-first CLI/TUI dashboard for tracking AI coding token costs across Claude Code, Cursor, Codex, and others.
- codeburn report -p today --refresh 60
Claude
- /clear — Wipes all conversation history for a hard reset; use at task boundaries when switching to something unrelated.
- /resume — Opens a session picker or jumps directly to a named past session to pick up where you left off.
- /btw — Asks an ephemeral side question without adding it to the conversation context.
- /goal — Sets a completion condition and keeps Claude working autonomously across turns until it is met, using a separate evaluator model to check after each turn.
- /plugin marketplace add <owner>/<repo> — Adds a skill marketplace from a GitHub repository.
- esc — Interrupts the current thinking or execution mid-response.
- esc esc — Opens a message picker to rewind the conversation to a previous point and branch from there.
- ~/.claude/settings.json
```
"permissions": {
    "allow": ["Bash(npm run *)", "Read(./src/**)"],
    "ask":   ["Bash(git push *)"],
    "deny":  ["Bash(rm *)", "Read(./.env*)"]
}
```

mattpocock/skills — A collection of reusable AI agent skills for common software engineering workflows.
- /grill-me — Intensively questions your plan until every decision branch is resolved, ensuring full alignment before implementation begins.
- /to-prd — Synthesizes the current conversation into a product requirements document and posts it as a GitHub issue.
- try out:
  - /tdd - Test-driven development with red-green-refactor loop.
  - /write-a-skill — Creates new agent skills with proper structure, progressive disclosure, and bundled resources.
caveman - makes agent talk like caveman — cuts ~75% of output tokens, keeps full technical accuracy. Brain still big. Mouth small.
openshift-eng/ai-helpers — Developer productivity plugins and commands for Claude Code and other AI assistants.
- /code-review:pr <pr-url-or-number> — Performs a comprehensive code quality review of a GitHub PR, checking test coverage, idiomatic patterns, DRY/SOLID principles, and build verification.
- /code-review:pre-commit-review - Performs a comprehensive code quality review of staged and unstaged changes before committing. Otherwise same as :pr
aro-ai-tools

CLAUDE.md - Identity & behavior rules specific to Claude; Define Claude’s role, tone, coding conventions, and persistent rules
CONTEXT.md - Session snapshot: current goals, active tasks, temporary tribal knowledge