pre-code-readiness-gate

Assesses a coding task for readiness before any code is written by running five weighted evidence checks — duplicate search, architecture fit, official documentation for the exact pinned release, open-source precedent, and root cause or gap — whose passing weights must total at least 0.90 out of 1.00, with lower totals producing a hold that asks the user to choose an approach or a stop that requests missing context. Use when a feature, bug fix, refactor, or integration is about to be coded, including changes that look like a two-line edit. Not for grading code that already exists, for research, prose, or question answering, and never as a substitute for tests or code review.

Version

v1.0.0

Size

18 KB

Updated

2026-10-12

Your client verifies the signature before reading. This explicit manual read is available on every plan. Free automatically loads five chosen skills; paid plans automatically load relevant skills across the library.