35 tagged defensive programming, measured the same way as everything else here.
Browse within: ai-assisted-development 22code-quality 22code-verification 22cpp 22error-prevention 13
Skill Claude CodeCodex
Software quality review and design guidance for any code: naming, cohesion, coupling, duplication, function size, dependency direction, and testability. Use when someone asks to improve, review, refactor, or design code, or asks what good structure looks like here. Routes to the sub-skill matching the kind of work.
Skill Claude CodeCodex
Defensive programming for any code: validating inputs at every boundary, checking for null and absent values, supplying safe fallbacks, catching and logging exceptions, and degrading gracefully rather than failing. Use when someone asks to make code robust, resilient, safer, or harder to crash. Routes to the sub-skill…
Skill Claude CodeCodex
AI features you ship to users: structured output, tool schemas, prompt injection, evals. Use when "the model returns bad JSON", "it hallucinates", "stop it calling the wrong tool", "add evals", or an LLM feature can trigger refunds, emails or writes. Covers schema-constrained output, idempotent tool calls…
Skill Claude CodeCodex
An isolated review role that checks whether work meets its acceptance criteria and reports its findings without changing the work's status.
Skill Claude CodeCodex
A system for recording runtime failures as structured “noise cards,” which are saved notes about unexpected behavior and its cause. Unresolved cards can later be turned into regression tests, tests that prevent the same failure from returning.
Skill Claude CodeCodex
A noise-card system for recording runtime failures as structured notes that can later become regression tests, which are tests that prevent the same problem from returning.