click

Click is a workflow guardrail for verification and reuse that leaves model selection and reasoning settings unchanged. It records each check's exact receipt, decides at execution time which checks must run again, and replaces a skipped run with one line instead of its full output. A check is skipped only when an exact same-state receipt, a signed input observation, or a committed owner policy proves it still holds; anything ambiguous runs. This is incremental verification, backed by revision-aware evidence: a record of what passed and whether it remains valid now. Every decision carries a reason code, and the dashboard shows where reuse was lost.

The tool targets repeated execution and repeated reading. It provides less waiting between edits by avoiding eligible unchanged checks while rerunning affected ones, and less context spent because a reused check returns one receipt line. It never makes a wrong skip, provides a reason for each decision, maintains continuity across tasks, and offers a visible outcome through a local dashboard showing verification, estimated avoided execution time, and output the host did not read again. Click fits projects with slow checks, repeated edit/test cycles, and separable test groups.

Click ships as a plugin for Codex CLI and Claude Code. Both hosts run the same runtime, evidence rules, and click-gate commands. It operates in three modes: Evidence (default), which records work and verification under host permissions; Guarded (opt-in), which stages a readable contract and waits for explicit approval; and Off, which leaves execution to the host without Click's workflow enforcement. The current release is v1.3.0.