claude

engineer-augmentor

Engineering co-pilot that keeps you the author: hypothesis-driven debugging coach, failing-test triage, logic audits before bugs ship, review-feedback planning, adversarial shell/PowerShell script review, pre-build specs, task decomposition, and client-side key exposure checks. Explains and checks; you change the code.

0 on GitHubby Kyle Rosebrookv1.4.1MITupdated todaydevelopment
/

Plugin engineer-augmentor from marketplace augmentor-packs in Krosebrook/augmentor-packs. Commands read from the repo's manifest — today.

What it installs

Skills (8)

  • client-key-exposure-check

    Use when the user builds a front-end, browser extension, mobile app, static site, or prototype that calls a paid or privileged API. Finds API keys and secrets exposed to the client, explains the risk, and outlines a fail-closed server-side proxy pattern with rate limits and origin checks. Reviews an

  • debugging-coach

    Use when the user is stuck on a bug, flaky behavior, or unexplained failure. Runs a hypothesis-driven investigation: reproduce, gather evidence, rank hypotheses, design the cheapest discriminating experiment, and confirm the root cause before any fix. Coaches the method; never guesses a fix as fact.

  • failing-test-triage

    Use when a test suite has failures and the user wants to understand and fix them efficiently. Groups failures by shared root cause, orders them by dependency, and fixes or proposes fixes one group at a time, deciding whether the test or the code is wrong. Never deletes or weakens tests to go green.

  • logic-audit

    Use when the user wants code checked for logical errors before they become bugs: control-flow gaps, unhandled states, off-by-one and boundary errors, data contract mismatches, race conditions, and silent failure paths. Reads and reasons; reports findings with evidence and test cases to prove them.

  • pre-build-spec

    Use before starting a feature, script, service, or refactor. Produces a short pre-build spec: problem, scope and non-goals, interfaces and data, edge cases, risks, test plan, rollout and rollback, and open questions. Interviews for missing facts and marks them; does not start coding until the user a

  • review-feedback-planner

    Use when the user receives code review comments or PR feedback and wants to process them systematically. Parses each comment, classifies it (bug, design, style, question, nit), proposes a response and change plan, flags disagreements to discuss, and checks nothing is missed. Drafts replies; the user

  • script-red-team

    Use when the user writes or reviews a Bash or PowerShell script, automation job, or one-liner that will run on real systems. Adversarially reviews for safety: quoting, error handling, idempotency, destructive operations, secrets handling, privilege, and portability, with a dry-run plan. Reviews and

  • task-decomposer

    Use when a request or project is large, multi-part, or ambiguous. Breaks it into workstreams and tasks with dependencies, identifies what can run in parallel, defines done for each, and plans the integration and verification step. The user assigns and executes.

Slash commands (4)

  • /engineer-augmentor:audit-logic

    Audit a file or diff for logical errors with test cases

  • /engineer-augmentor:debug

    Hypothesis-driven debugging session for a stuck bug

  • /engineer-augmentor:review-plan

    Turn PR review comments into a change plan and reply drafts

  • /engineer-augmentor:spec

    Write a pre-build spec before coding

Subagents (2)

  • logic-auditor

    Read-only logic auditor. Reads the specified files or recent diff and reports logical errors (boundaries, unhandled states, data contract mismatches, races, swallowed errors) with file:line evidence and a test case per finding. Use before merging risky changes.

  • script-reviewer

    Read-only adversarial reviewer for Bash and PowerShell scripts. Flags unquoted expansions, unsafe deletes, missing error handling, non-idempotent steps, hard-coded secrets, and privilege issues, and runs shellcheck/PSScriptAnalyzer if available. Use before running a script on real systems.

View source on GitHub

More from Krosebrook/augmentor-packs

All 20 plugins →

For people with some technical background moving into building with AI: prompt engineering, agentic workflow design, LLM APIs, RAG, evals, shipping a prototype, a learning roadmap, reviewing AI-generated code, and LLM app security.

9 skills · 5 commands · 2 agents

0

Co-pilot for parents and family caregivers: family logistics and meal planning, a homework coach that doesn't give answers, school and doctor message drafts, aging-parent appointment prep, a care log, benefits/legal question checklists that defer to professionals, and self-care.

9 skills · 5 commands · 2 agents

0

Sparring partner for solo founders: positioning and offer, customer discovery scripts and synthesis, pricing, landing page copy, outreach drafts, weekly CEO review, runway math, SOPs/delegation, and a decision journal.

9 skills · 5 commands · 2 agents

0

Start here if you're new to AI: a warm, jargon-free first-week tour, asking clear questions, checking answers and spotting made-up facts, what not to share, picking the right pack, breaking messy tasks into steps, practice with feedback, and a plain-words glossary. Coaches; never replaces you.

8 skills · 5 commands · 2 agents

0

Planning and feedback co-pilot for teachers: standards-aligned lesson and unit plans (you supply the standards), differentiation/UDL, rubrics, assessments, feedback suggestions on student work (you stay the grader), parent communication, and classroom AI policy.

8 skills · 5 commands · 2 agents

0

Knowledge-work co-pilot: inbox triage and reply drafts, meeting prep and notes-to-actions, document/report drafting, spreadsheet formulas and analysis, slide outlines, stakeholder updates, SOPs, and polite-pushback/negotiation coaching. Drafts only; you send.

8 skills · 5 commands · 2 agents

0

Similar plugins

Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).

39 skills

mattpocock/skills 323,108

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.

5 commands

HKUDS/CLI-Anything 58,665

Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer

8 skills · 1 MCP server

ChromeDevTools/chrome-devtools-mcp 55,181

Automates browser interactions for web testing, form filling, screenshots, and data extraction

12 skills

vercel-labs/agent-browser 50,373

Maintain engineer-augmentor?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins20 plugins