Code review, debugging, architecture, testing and docs. 100 reviewed prompts for engineering.
Requires real evidence before a software change is reported as finished: build and tests run, the change exercised directly, at least one edge case tried, and results shown. Use when you implement, fix or verify code and can run checks.
Reviews how an organization stores and shares sensitive data and proposes a prioritized protection plan covering access, encryption, sharing rules and monitoring. Use when tightening data handling in a small team.
Use when you have just written or changed code and before presenting it: runs a focused second pass for the mistakes generated code commonly contains, fixes what it can, and reports what was fixed and what needs the author's attention.
Use when you are about to commit a non-trivial technical or business decision, claim something is safe or correct, or change something hard to undo. Runs a short, bounded adversarial review of the decision before it stands.
Designs a tested, low-noise detection rule for a suspicious behavior in your own logs, with false-positive tuning, evasion considerations, a safe test plan and analyst triage steps. Use it for defensive security monitoring.
Guides a bug fix that starts with a failing test, then the smallest fix, then proof nothing else broke. Use it when you or an assistant must fix a defect without causing regressions.
Produces an evidence-based debt register for a codebase or module: ranked findings with file references, severity and effort, top fixes, quick wins and things that only look bad. Use it to plan refactoring work.
Designs a firmware module for a resource-limited microcontroller: stated hardware assumptions, task and interrupt layout, memory and timing rules, a C code skeleton, and a verification and rollout plan. Use when starting a driver or feature on embedded hardware.
Produces a cloud architecture proposal for a workload with components, trade-offs, cost drivers, security and reliability notes, and a migration path. Use it when designing a new system or reviewing an existing setup.
Designs a reliable connection between two systems: data flow, pattern choice, authentication, retries, duplicate protection, monitoring and a build plan. Use when planning a new integration or fixing a fragile one.
Reviews an AI agent or automation design for human control, handling of unclear instructions, prompt-injection defenses, transparency and data exposure, and names concrete fixes. Use it before giving an agent real tools or data.
Diagnoses a slow endpoint, job or page from your symptoms and measurements, and returns likely root causes, a ranked fix list with expected gains and a before/after measurement plan. Use it before scaling hardware.
