What problem does it solve? After an implementation pass, it is hard to know whether the code actually satisfies everything the feature's spec, plan, and tasks promised. This Skill closes that gap by auditing the current codebase against those artifacts and appending any unfinished work as new, traceable tasks so the implement step can finish the job. ## Core Features & Use Cases - Gap Assessment: Reads spec.md, plan.md, and tasks.md as the source of intent, inspects the in-scope code, and classifies findings as missing, partial, contradicts, or unrequested. - Severity Grading: Ranks findings CRITICAL through LOW, treating constitution MUST violations as the highest severity. - Append-Only Task Generation: Appends a new numbered Convergence phase to tasks.md with zero-padded task IDs, never rewriting or renumbering existing tasks. - Use Case: After running the implement command on a feature, run converge to discover that an acceptance criterion (e.g., an append-only guard) was never built; it appends a task like "T042 Add append-only enforcement per FR-008 (missing)" for the next implement pass. ## Quick Start Ask the agent to run the speckit-converge skill to check the current feature's implementation against its spec and append any remaining work to tasks.md.
