Home / Subagents / cavecrew-reviewer

cavecrew-reviewer

codejunkie99/agentic-stack-desktopSubagent

Review a diff, branch, or file for bugs and regressions. Use proactively after any edit and before commit or PR so the parent never spends Fable tokens re-reading diffs. Grok 4.6, read-only, caveman output.

ScannedSubagentSource: GitHub
76 stars Updated 1mo ago License Apache-2.0 Context β‰ˆ56 tokens in your chat
58popular

Why this score

Stars10/35
Usage (forks)3/15
Upkeep20/25
License15/15
Owner history10/10

This repo hosts 3 subagents, so its 76 stars are split between them (about 25 each) for scoring.

Popularity score: 58/100 from public GitHub signals only. It shows how widely used and maintained this is, not how well it works. Not included yet: the fake-star check. The security scan is shown separately and doesn't change the score. How we rank

What it can use

  • Every tool, because it sets no tools list (it inherits all of yours, including shell and MCP tools)

Model: grok-4.6

This agent can act outside your files. Read it before installing, and narrow its tools: line if it needs less.

The agent file

---
name: cavecrew-reviewer
description: Review a diff, branch, or file for bugs and regressions. Use proactively after any edit and before commit or PR so the parent never spends Fable tokens re-reading diffs. Grok 4.6, read-only, caveman output.
model: grok-4.6
readonly: true
---

You are cavecrew-reviewer. Diff reviewer on Grok 4.6. Read-only. Never edit files. Never run mutating commands.

Input is a `git diff` range, a branch, or paths. Read that slice. Hunt bugs and regressions. Do not redesign.

Look for: logic bugs, broken references, frontmatter/YAML errors, missing newline at EOF, secrets, scope creep beyond the stated task.

Findings only. No architecture opinions. No praise.

Output contract (verbatim):
```
path:line: <emoji> <severity>: <problem>. <fix>.
totals: NπŸ”΄ N🟑 NπŸ”΅ N❓
```
or `No issues.` Sorted file then line ascending.

Severity: πŸ”΄ blocker, 🟑 should-fix, πŸ”΅ nit, ❓ needs-human.

Auto-clarity: security findings in plain English so the parent cannot misread them. Resume caveman after.

Caveman full otherwise. Never set `model: inherit`.

Apache-2.0 license, from codejunkie99/agentic-stack-desktop at commit c9a02882ea60.

Security

Scanned Oct 21, 2026 at commit c9a02882ea60, with Cisco's open-source skill-scanner.

None of the patterns we list turned up: shell commands, network requests, reading environment variables, downloading and running code, or installing packages.

A scan lists patterns in the code. It can't prove an add-on is harmless, and it can miss things, so read what you install. How we scan

Alternatives

Other makers' add-ons that do the same job, next to this one.

NameScoreStarsContextUpdated
This onecavecrew-reviewer codejunkie99/agentic-stack-desktop Β· Subagent 58β˜… 76β‰ˆ56 in chat1mo ago
cavecrew JuliusBrussee/caveman Β· Skill Both about cavecrew 90β˜… 114kβ‰ˆ22 always16d ago
reviewer Goldziher/ai-rulez Β· Subagent Both about reviewer and diff 56β˜… 163β‰ˆ6 in chat16d ago
copilot-reviewer Waishnav/devspace Β· Subagent Both about reviewer and regression 77β˜… 5.3kβ‰ˆ24 in chat23d ago
referee2-reviewer flonat/flonat-research Β· Subagent Both about reviewer 51β˜… 147β‰ˆ23 in chat1mo ago
atomic-reviewer damusix/atomic-claude Β· Subagent Both about reviewer and branch 58β˜… 114β‰ˆ28 in chat16d ago