caveman-optimize
Evaluate Caveman Optimization Candidates
Aggregate repository profiles do not identify a safe or valuable code change by themselves. This skill tests one approved candidate against a controlled baseline.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Review the Skillstore skill "caveman-optimize" from https://skillstore.io/skills/juliusbrussee-caveman-optimize.md and its manifest at https://skillstore.io/api/skills/juliusbrussee-caveman-optimize/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-optimize". Show the current optimization observations.
Expected outcome:
- The skill lists each supported observation with its exact server title and recorded profile.
- It asks the operator to select one item before repository inspection.
Using "caveman-optimize". Evaluate my selected tool-output-size profile.
Expected outcome:
- The report identifies one evidenced callsite and the approved minimal change.
- It compares baseline and candidate results on the same fixture and measurement.
- It records quality checks, code checks, accounting limits, and the final decision.
Security Audit
SafeAll 20 static findings are false positives caused by Markdown code fences, inline code identifiers, or ordinary prose. The only documented command is a fixed read-only Caveman CLI query, and the workflow requires operator approval before repository changes.
Risk Factors
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-optimize/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/juliusbrussee-caveman-optimize?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/juliusbrussee-caveman-optimize?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/juliusbrussee-caveman-optimize/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/juliusbrussee-caveman-optimize.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-optimize security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/juliusbrussee-caveman-optimize/audits/1BibTeX citation
@techreport{juliusbrussee-juliusbrussee-caveman-optimize-2026,
author = {juliusbrussee},
title = {caveman-optimize security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/juliusbrussee-caveman-optimize/audits/1},
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-optimize security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
- name: "juliusbrussee"
date-released: "2026-08-12"
url: "https://skillstore.io/skills/juliusbrussee-caveman-optimize/audits/1"
identifiers:
- type: other
value: "skillstore:juliusbrussee-caveman-optimize:audit:1"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: MediumWhat You Can Build
Reduce Context Window Load
Evaluate a focused code change against a current context-window profile using identical baseline and candidate inputs.
Control Tool Output Size
Test whether a specific output reduction preserves task quality while improving the selected resource measure.
Review Optimization Evidence
Verify that a proposed optimization has callsite evidence, controlled measurements, and an explicit operator decision.
Try These Prompts
Read the current Caveman report-only observations. Show supported items with their exact identifiers, titles, observations, and last-seen times. Do not rank them.
I choose [observation id]. Inspect the repository for one evidenced callsite that could produce this profile. Do not edit any files yet.
For [observation id], propose one minimal candidate and a paired baseline evaluation. Specify the fixture, quality check, resource measure, commands, environment, and confounders.
Apply the approved candidate for [observation id]. Run the paired evaluation and focused checks. Keep it only if quality and resource evidence support adoption.
Best Practices
- Refresh observations through the authenticated CLI before every evaluation.
- Use identical fixed inputs and the same resource measurement for both evaluation arms.
- Approve the candidate and evaluation design before editing repository files.
Avoid
- Do not infer dollars, savings, or production impact from aggregate profiles.
- Do not use retired identifiers or historical proposals as current evidence.
- Do not keep a candidate when quality regresses or the comparison is inconclusive.
Frequently Asked Questions
Does this skill automatically optimize my repository?
Which observations are supported?
Does it estimate financial savings?
What happens when Caveman authentication fails?
Can ordinary unit tests prove the optimization?
Can the skill change Caveman opportunity state?
Developer Details
Author
juliusbrusseeLicense
MIT
Skillstore revision
r1
Version notice
The author did not declare a version.
Ref
9e5d158a7fd6576779833ff3529fe92b1a198150
Maintenance freshness
8/13/2026
Usage
0 downloads · 0 views
File structure
📄 SKILL.md