All shelves / Mods / Workflow
allowlist-coach
By ice-lfernandes
allowlist-coach is a Claude Code workflow mod by ice-lfernandes.
Counts permission dialogs you answer and, after repeated approvals, offers that rule for the allow list, asking you before it writes.
Draws
Install
The marketplace name is lfernandes-mods.
/plugin install allowlist-coach --marketplace ice-lfernandes/claude-code-modsIt runs inside the Claude Code process. The side panel is only what the author's README names. Read the repo before you install, or run claude plugin validate on the plugin folder.
- Version
- 0.3.3
- License
- MIT
- Repo stars
- 7
- Last change
- 2026-10-11
What it can reach
- Draws in the session
- Sees tool calls
The repository README says these mods require Claude Code 2.1.287 or later. Under a permission dialog it draws how many times you approved that rule in this project, and it toasts once when a rule is ready. /allowlist opens a pane of every rule, with its approvals, refusals, and the last call that asked. allow asks first and shows the line it would add. The first answer is Not now, so a stray Enter writes nothing. You can add the rule to .claude/settings.local.json or to .claude/settings.json, or never offer it. remove asks first, with Cancel first. The pane's allow, dismiss, remove, and reset buttons put the command in the prompt for you to finish and send. A rule you refused, a whole tool, a broad wildcard, and a command that deletes, escalates, reaches the network, or runs arbitrary code is shown as risky and is not offered. A call that carries a credential is not counted, so its command is never stored. Dialogs a settings hook answered, and decisions in auto, dontAsk, or bypassPermissions, are not counted. threshold defaults to 5. language defaults to auto, Portuguese when the system language is Portuguese, English otherwise. It writes a plain file in the project's .claude folder only after you confirm. The reach table marks network, processes, model calls, and sending data anywhere as no.
The install line has no @suffix. The marketplace name is lfernandes-mods. The marketplace also holds limits-meter, agent-watch, and plain-view. This one installs on its own. The count under the dialog, and the pane of rules, are why this page files it under workflow. command-explainer draws a plain-English line and a risk level on the permission prompt, so you can see what the command does before you approve it. Nothing else on the shelf counts those answers and offers an allow rule after you confirm. Risky rules stay listed and are not offered.
Demo
What it hooks into
session.startWhen a session startstool.callAround a tool call, where it can hold oneclassic.PermissionRequestA Claude Code eventclassic.PostToolUseA Claude Code eventclassic.PostToolUseFailureA Claude Code eventcommand.runWhen you run its slash commandui.scrollA Claude Code eventPaneui.renderDraws into the sessionPane
Read from the on() calls in its hook modules. Open its source under Files to check what each one does.
Files
The author's README and source, 4 files. They load when you open them.
Facts
- Author
- ice-lfernandes
- Category
- Workflow
- Marketplace
- lfernandes-mods
- Folder
- allowlist-coach
- Version
- 0.3.3
- License
- MIT
- Repo stars
- 7
- Last change
- 2026-10-11
- Language
- TypeScript
- Files
- 11