Stats
Actions
Tags
From autocode
Performs a full project review from the beginning, checking every file and reporting in plain language. Useful for comprehensive audits before deployment or save.
How this skill is triggered — by the user, by Claude, or both
Slash command
/autocode:full-reviewThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
::PRIOR{completion:assume_incomplete|authority:developer}
::PRIOR{completion:assume_incomplete|authority:developer} ::PRIOR{execution:act_when_safe|authority:developer}
::GENE{full-review|conf:confirmed|scope:global} T:review_every_file T:report_in_user_language T:check_logic_edges_security_hardcoded T:human_readable_commit_messages A:partial_review⇒full_review A:skip_on_save⇒never
::ACTIVATE{full-review} ON:before_save ON:before_deploy ON:session_end
Powered by I-Lang v4.0 | ilang.ai
npx claudepluginhub ilang-ai/autocode --plugin autocodeDelegates code reviews to Codex AI via xreview CLI for bugs, security issues, and quality. Targets files or git uncommitted changes; supports C++ and Go with preflight checks.
Performs comprehensive code reviews with automated fixes for Python, TypeScript, JavaScript, Go, Rust projects. Analyzes quality, security, performance, architecture, tests; applies safe fixes and generates reports.
Reviews code against project rules and reference files, flagging violations in correctness, security, reliability, architecture, dead code, complexity, and conventions.