Skip to content
Back to skills

Review Writes

ASecurity

Pre-approves push-branch.sh, reply-resolve.sh and react-marker.sh for the rest of the invoking turn and runs no command. Invoke right before the first such call.

  • 2 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 18, 2026
ai-agentsshellbashreactgit

Security analysis

A100/100

Scanned November 18, 2026

npx -y skills add brenpike/hivemind --skill review-writes --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Review Writes?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Review Writes
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/brenpike-review-writes/badge)](https://www.skillsdirectory.com/skills/brenpike-review-writes)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: review-writes
description: Pre-approves push-branch.sh, reply-resolve.sh and react-marker.sh for the rest of the invoking turn and runs no command. Invoke right before the first such call.
user-invocable: false
allowed-tools:
  - Bash(bash "${CLAUDE_PLUGIN_ROOT}/skills/push-branch/scripts/push-branch.sh" *)
  - Bash(bash "${CLAUDE_PLUGIN_ROOT}/skills/github-review-loop/scripts/reply-resolve.sh" *)
  - Bash(bash "${CLAUDE_PLUGIN_ROOT}/skills/github-review-loop/scripts/react-marker.sh" *)
shell: bash
---

This skill runs no command.

Its allowed-tools pre-approve exactly three scripts, each by its plugin-root path, for the rest of the turn that invokes it:

- `${CLAUDE_PLUGIN_ROOT}/skills/push-branch/scripts/push-branch.sh`
- `${CLAUDE_PLUGIN_ROOT}/skills/github-review-loop/scripts/reply-resolve.sh`
- `${CLAUDE_PLUGIN_ROOT}/skills/github-review-loop/scripts/react-marker.sh`

No settings rule is needed for these calls.

Invoke it right before the first such call, after any other skill call. No other skill invocation may come between the grant and the writes.

The caller runs each script in its own documented form. This skill restates none of them.

Deny and ask rules still win. A refused call is the caller's to handle.

This completes this skill's own procedure. The skill then returns control to whatever invoked it; the caller then continues from the point at which it invoked this skill.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in ai-agents →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.