← all publishers

habonn

@habonn source repo

11 published skills

  1. Tdd · habonn
    Test-driven development with red-green-refactor loop. Builds features or fixes bugs one vertical slice at a time. USE FOR: building features with TDD, red-green-refactor, writing integration tests, test-first development, fixing bugs with tests. DO NOT USE FOR: generating bulk test files (use test-specific skills), architecture review, debugging without tests.
    0
    installs
  2. Caveman · habonn
    Ultra-compressed communication mode. Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. USE FOR: saving tokens, brief responses, terse mode, less verbose output. DO NOT USE FOR: documentation writing, user-facing content, onboarding.
    0
    installs
  3. Diagnose · habonn
    Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. USE FOR: debugging, diagnosing bugs, performance regressions, broken behavior, failing tests, error investigation. DO NOT USE FOR: writing new features, architecture review, code style.
    0
    installs
  4. Grill Me · habonn
    Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. USE FOR: stress-testing plans, design reviews, decision-making, getting grilled on architecture or feature design. DO NOT USE FOR: code-specific grilling with docs (use grill-with-docs), writing code, debugging.
    0
    installs
  5. Zoom Out · habonn
    Tell the agent to zoom out and give broader context or a higher-level perspective on an unfamiliar section of code. USE FOR: understanding unfamiliar code, getting a map of modules and callers, seeing the bigger picture, onboarding to a new area. DO NOT USE FOR: writing code, debugging, making changes.
    0
    installs
  6. Commit Msg · habonn
    Smart git commit workflow using Conventional Commits format with AI-generated commit message suggestions based on staged changes. USE FOR: writing commit messages, staging changes, conventional commits, atomic commits, breaking changes. DO NOT USE FOR: git branching, merging, rebasing, CI/CD pipelines.
    0
    installs
  7. Skill Auditor · habonn
    Analyze repository and generate custom SKILL.md with project-specific tech stack, architecture, and conventions. USE FOR: auditing repos, generating skill files, detecting architecture patterns, understanding project structure after major refactors. DO NOT USE FOR: writing application code, running tests, deploying.
    0
    installs
  8. Grill With Docs · habonn
    Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. USE FOR: stress-testing plans against project language, building shared vocabulary, documenting decisions, domain-driven design sessions. DO NOT USE FOR: non-code grilling (use grill-me), writing code, debugging.
    0
    installs
  9. Daily Commit Summary · habonn
    Generate daily task summaries by analyzing git commits. MUST include analyzed Tasks section. USE FOR: daily standups, end-of-day summaries, commit analysis, task transformation from raw commits. DO NOT USE FOR: sprint summaries (use sprint-commit-summary), writing commits (use commit-msg).
    0
    installs
  10. Sprint Commit Summary · habonn
    Generate sprint commit summaries by analyzing git commits across a 2-week sprint period. Supports date range input like "6-17" for quick sprint specification. USE FOR: sprint reviews, demo prep, 2-week summaries, sprint statistics. DO NOT USE FOR: daily summaries (use daily-commit-summary), writing commits (use commit-msg).
    0
    installs
  11. Improve Codebase Architecture · habonn
    Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. USE FOR: improving architecture, finding refactoring opportunities, consolidating tightly-coupled modules, making code more testable and AI-navigable. DO NOT USE FOR: writing new features, debugging, quick fixes.
    0
    installs