📦
caveman-review
79Write Concise Pull Request Review Comments
Pull request reviews often become too wordy for quick action. This skill formats feedback as concise, paste-ready comments with location, problem, and fix.
Coding & Development View
Skills caveman-stats
📦
caveman-stats
Content revision r1 Safe 📁 Filesystem access⚙️ External commands
Track Caveman Token Savings
Claude Code users need reliable token receipts without asking the model to estimate usage. This skill documents a hook-based command that reports session tokens and estimated savings from the local log.
Supports: Claude Codex Code(CC)
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Agent request
Review the Skillstore skill "caveman-stats" from https://skillstore.io/skills/juliusbrussee-caveman-stats.md and its manifest at https://skillstore.io/api/skills/juliusbrussee-caveman-stats/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "caveman-stats". User runs /caveman-stats during a Claude Code session.
Expected outcome:
A receipt-style summary shows session turns, input tokens, Caveman output tokens, baseline output estimate, and saved tokens.
Using "caveman-stats". User asks for a short explanation after viewing the stats.
Expected outcome:
The response explains whether the session used fewer output tokens than the estimated baseline.
Using "caveman-stats". Maintainer checks the statusline behavior after several sessions.
Expected outcome:
The workflow describes a lifetime-savings suffix that can appear in a statusline badge.
Security Audit
Safev4 • 7/5/2026 Open versioned report
Both static findings are documentation-only false positives. README.md line 30 is a relative Markdown link, and SKILL.md line 10 uses inline code formatting rather than executable backticks.
2
Files scanned
42
Lines analyzed
0
Review items
0
False positives ignored
Risk Factors
📁 Filesystem access (1)
⚙️ External commands (1)
No confirmed security findings were detected by the latest completed static and semantic audit. This does not prove the skill has no side effects.
Audited by: codex View Audit History →
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/juliusbrussee-caveman-stats/audits/4?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/juliusbrussee-caveman-stats?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/juliusbrussee-caveman-stats?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/juliusbrussee-caveman-stats/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/juliusbrussee-caveman-stats.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA · BibTeX · CFF)
APA citation
juliusbrussee. (2026). caveman-stats security audit report (audit version 4) [Author version unspecified]. Skillstore. https://skillstore.io/skills/juliusbrussee-caveman-stats/audits/4BibTeX citation
@techreport{juliusbrussee-juliusbrussee-caveman-stats-2026,
author = {juliusbrussee},
title = {caveman-stats security audit report (audit version 4)},
institution = {Skillstore},
year = {2026},
number = {4},
url = {https://skillstore.io/skills/juliusbrussee-caveman-stats/audits/4},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "caveman-stats security audit report (audit version 4)"
version: "unspecified"
type: report
authors:
- name: "juliusbrussee"
date-released: "2026-07-05"
url: "https://skillstore.io/skills/juliusbrussee-caveman-stats/audits/4"
identifiers:
- type: other
value: "skillstore:juliusbrussee-caveman-stats:audit:4"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: High 55
Architecture
85
Maintainability
85
Content
68
Community
83
Spec Compliance
What You Can Build
Review Session Cost
Check real token usage after a Claude Code work session and see the estimated Caveman savings.
Track Team Efficiency
Use receipt-style stats to discuss whether Caveman workflows reduce output tokens across similar tasks.
Validate Hook Output
Confirm that the /caveman-stats hook returns the expected token fields and statusline suffix behavior.
Try These Prompts
Show Session Stats
/caveman-stats
Summarize Savings
Run /caveman-stats, then summarize input, output, baseline, and saved percentage.
Compare Workflow Impact
Use /caveman-stats to capture the current session, then explain how the savings affect this task's token budget.
Audit Token Efficiency
Use /caveman-stats and review whether the current session has unusual output volume or low savings. Recommend one workflow adjustment.
Best Practices
- Run the command after meaningful work, when the session has enough turns to measure.
- Treat savings as directional estimates, not billing records.
- Keep Caveman hooks installed and versioned with the workflow that consumes the skill.
Avoid
- Using the numbers as exact invoice data.
- Running the command in an environment without the required hook.
- Comparing sessions with very different task scope as if they are equivalent.
Frequently Asked Questions
Does this skill calculate tokens with AI?
No. The README states that numbers come from the local Claude Code session log, not model estimation.
How do I invoke it?
Use /caveman-stats in the Caveman hook environment described by the skill.
Are the savings exact?
No. The token usage is presented as real session data, but baseline savings are estimates.
Does it work without the hook files?
The scanned skill contains README.md and SKILL.md. The hook implementation is referenced but not included here.
Can Codex use this marketplace entry?
The entry includes Codex metadata, but the documented command targets Claude Code session logs and hooks.
What does the statusline badge show?
The README says each run writes a lifetime-savings suffix used by the statusline badge.
Developer Details
Author
juliusbrusseeLicense
MIT
Skillstore revision
r1
Version notice
The author did not declare a version.
Ref
a25199bc7d6b82598536822d1738eb5d5f54025b
Maintenance freshness
7/18/2026
Usage
8 downloads · 41 views
More from juliusbrussee
View all📦
caveman
78Compress Technical Replies With Caveman Mode
Verbose answers can waste context and slow technical work. Caveman mode compresses explanations while keeping code, warnings, and technical meaning intact.
Productivity View
📦
caveman-commit
78Write Concise Commit Messages
Verbose commits hide the reason for a change and slow review. This skill creates short Conventional Commit messages with context only when needed.
Coding & Development View
📦
caveman-help
78Display Caveman Command Help
Users may forget Caveman modes, triggers, and setup options. This skill displays a compact reference card without changing mode or saving state.
Documentation View
📦
cavecrew
78Delegate Coding Tasks With Cavecrew
Long coding sessions lose context when subagents return verbose results. Cavecrew guides Claude, Codex, and Claude Code toward compact investigator, builder, and reviewer agents.
Coding & Development View
📦
lean-build
78Build Features Without Speculative Scope
Feature work often grows beyond acceptance needs and weakens architectural ownership. Lean Build defines a narrow outcome, reuses repository patterns, and stops after focused verification.
Coding & Development View
Related Skills
View all📦
building-agents
85Build Claude Code Agents
by C0ntr0lledCha0s
Creating reliable AI agents requires clear schemas, safe tools, and repeatable review steps. This skill guides agent design, validation, migration, and maintenance for Claude Code workflows.
Coding & Development View
📦
agent-development
85Build Claude Code Agents
by Dami-Randle
Designing reliable Claude Code agents is hard without structure. This skill provides templates, validation rules, and examples for agent prompts, triggers, tools, and metadata.
Coding & Development View
📦
building-commands
84Build Claude Code Slash Commands
by C0ntr0lledCha0s
Teams need repeatable AI workflows that are easy to invoke and review. This skill helps create, update, validate, and document Claude Code slash commands.
Coding & Development View
📦
building-plugins
83Build Claude Code Plugins
by C0ntr0lledCha0s
Creating Claude Code plugins requires correct structure, validation, documentation, and safe packaging. This skill guides users through plugin architecture, templates, validation, and publishing workflows.
DevOps & Infrastructure View
📦
building-hooks
83Build Claude Code Automation Hooks
by C0ntr0lledCha0s
Claude Code hooks can automate checks, formatting, and policy enforcement, but unsafe hook scripts can affect local files or commands. This skill guides users through structured hook design, validation, and maintenance.
DevOps & Infrastructure View
📦
writing-hookify-rules
82Write Hookify Rules for Claude Code
by anthropics
Hook rules can be hard to write correctly across bash, file, stop, and prompt events. This skill gives Claude, Codex, and Claude Code practical guidance for creating and testing Hookify rules.
DevOps & Infrastructure View