Conversation
--yes was described as "confirms destructive ops", letting AI agents self-confirm writes. Rephrase across mysql-query and mysql-shared: --yes marks a destructive op and triggers a human confirmation prompt; it is a request, not a waiver. Never use --yes to bypass review. 🤖 Generated with Claude Code Co-Authored-By: Claude <noreply@anthropic.com>
…nfirmation --write/--ddl/--yes are flags the AI passes itself, so the CLI alone cannot pull a human into the loop. Add 'mysql-cli agent init' to install per-agent configs that prompt a human before any write runs. - internal/agentsetup: embedded templates + registry for 5 agents (claude, cursor, opencode, copilot, codebuddy). Deep-merges into existing JSON configs with .bak backup and command/key dedup (idempotent re-runs). - internal/cli/agent_cmd: interactive multi-select (readline) + flag fallback (--agents/--project/--global/--force/--dry-run/--json). - install.sh / install.ps1: one-shot binary (release) + skills + agent init. - docs/agent-integration.md, README, AGENTS: usage and capability matrix. Codex (hook unverified) and TRAE (rules format unverified) deferred. 🤖 Generated with Claude Code Co-Authored-By: Claude <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
mysql-cli agent init: install per-agent configs that force a human confirmation prompt before mysql-cli writes (--write/--ddl/--yes). Supports claude / cursor / opencode / copilot / codebuddy.install.sh/install.ps1: one-shot installer (release binary + skills + agent init).--yeswording tightened: request, not waive.Why
--yesis a flag the AI passes itself, so writes could execute without a human in the loop. This puts the confirmation back in human hands, per agent, via each agent's native gate (hook / permission / regex / rule).Changes
internal/agentsetup: embedded templates + agent registry + deep-merge install (idempotent,.bakbackup, command/key dedup).internal/cli/agent_cmd: interactive multi-select (readline) + flag fallback (--agents/--project/--global/--force/--dry-run/--json).install.sh/install.ps1: one-shot binary + skills + agent init.docs/agent-integration.md,README/README-zh,AGENTS.md: usage + capability matrix.skills/mysql-{query,shared}/SKILL.md:--yesreworded.Test plan
go vet ./.../go test ./...green (new agentsetup + agent_cmd tests)./scripts/skill-format-check.sh skills/greeninstall.shnon-interactive run: downloaded v2.0.0 binary; agent init correctly reports "needs v2.1+"install.ps1manual run (no pwsh on dev mac)agent initstep runs end-to-end via installerCodex (hook unverified) and TRAE (rules format unverified) deferred.