candid

Use when you have an output — copy, a doc, an answer, a plan, a prompt, any text artifact — and want it made better against your goal. Runs a critique→refine loop with configurable stop conditions and auto/review-each/interactive modes. Default is a single pass using the current session context to decide what "better" means.

Fast ship for low-risk changes — runs only the steps you explicitly enable in fastShip config

Use when the code works and you want a focused pass on improving approach, clarity, and quality — distinct from candid-review's defect hunt. Each suggestion includes a concrete before/after, tradeoffs, and confidence level.

Run candid-review in a loop until all issues are resolved, with configurable auto, review-each, or interactive modes and support for ignored issues

Use when reviewing code changes before commit or PR — configurable harsh or constructive review against Technical.md standards, with categorized issues, actionable fixes, todo tracking, and optional auto-commit.

Ship your changes - review, build, test, create PR, and optionally auto-merge

Take a candid-chrome-qa findings JSON, let the user pick which findings to fix, and dispatch fixes — single batched PR, one PR per finding via Conductor deep links, local-only, or issues-only. Optionally files Linear (or other tracker) issues per finding before shipping.

Generate Technical.md and config.json by deeply analyzing your codebase structure, architecture, and patterns

Validate Technical.md for vague rules, linter overlaps, and effectiveness issues

Drive a real Chrome session against a running web app to find bugs and write structured findings JSON. Use when the user asks for a QA pass, smoke test, UX audit, accessibility check, or "find bugs / polish / a11y / perf / copy issues" on a route.

Audit and optimize the context candid loads during reviews — Technical.md efficiency, exclude patterns, decision register, and config tuning