Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add JuliusBrussee/caveman-code --skill cavekit-revisiongit clone --depth 1 https://github.com/JuliusBrussee/caveman-codeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/juliusbrussee/caveman-code/cavekit-revision)<a href="https://agentmods.dev/skills/juliusbrussee/caveman-code/cavekit-revision"><img src="https://agentmods.dev/badge/skills/juliusbrussee/caveman-code/cavekit-revision.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00073 | $0.03211 |
| Opus 5 | $0.00036 | $0.01605 |
| Sonnet 5 | $0.00015 | $0.00642 |
| Haiku 4.5 | $0.00007 | $0.00321 |
Grade A, and why
cavekit-revision scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 344 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Revision: Tracing Bugs Back to Kits
In Cavekit, revision means tracing a production defect upstream through the cavekit chain until you find the gap that allowed it. In practice, when the built software has bugs or gaps, you trace the issue back to the kits and prompts and fix at the source -- not just in code.
Key insight: When a fix lives only in code with no corresponding cavekit update, the next iteration loop may reintroduce the same defect. The goal is that kits plus the iteration loop can reproduce any fix autonomously.
1. Why Revision Matters
Without revision, every bug fix is a one-off patch. The next time the iteration loop runs, it may reintroduce the bug because nothing in the kits or plans prevents it.
With revision:
- Bug fixes become cavekit improvements that persist across all future iterations
- The iteration loop becomes self-correcting -- it learns from every manual intervention
- Kits become progressively more complete over time
- The gap between "what kits describe" and "what works" shrinks monotonically
Without revision:
Bug found -> Fix code -> Bug may return next iteration
With revision:
Bug found -> Fix code -> Update cavekit -> Re-run iteration loop -> Fix emerges from kits alone
2. The 6-Step Revision Process
This is the complete process for tracing a bug back to its cavekit-level root cause and closing the loop.
Step 1: Identify and Fix the Defect
Locate the bug -- whether through manual testing, automated failures, user reports, or monitoring alerts -- and resolve it through normal debugging. This produces a working code change, but the job is far from done: until the underlying cavekit gap is closed, this fix is fragile.
# The fix produces commits that we will analyze
git log --oneline -5
# a1b2c3d Fix: connection pool exhaustion under concurrent load
# e4f5g6h Fix: missing rate limit headers in API responses
Step 2: Analyze What the Cavekit Missed
This is the pivotal step. Ask: "Where in the cavekit chain did this requirement slip through?"
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 344 lines · 73 tokens per session scan A 358aef0f621d
cavekit-revision is a skill published in the GitHub repository JuliusBrussee/caveman-code (932 stars, last pushed 23d ago), licensed MIT. It adds 73 tokens to every session and 3,211 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
systematic-debugging
Use when debugging a failing test, build error, or runtime issue that isn't immediately obvious. Guides a 4-phase root cause analysis instead of random fix attempts.
diagnose
Trace from a reproduced symptom to the source code that causes it. Pin the specific file and approximate line, rate confidence in the cause and clarity of the fix independently, and always propose a concrete fix.
repro-admin
Reproduce an EmDash admin UI bug. Attach a container, start the demo dev server, drive the admin with agent-browser using the dev-bypass session, and capture the reproduction as screenshots plus a replayable transcript.
log-error-digest
Analyze log files to troubleshoot errors, identify peak error periods, and produce error clustering, frequency statistics, and time distribution reports. Supports JSON, syslog, and Nginx formats with automatic detection. Use when a user uploads a .log file and asks to analyze errors, find patterns, debug issues, or…
byted-util-volcengine-detect-retry
An orchestration workflow for Volcengine Cloud Detect, a service that checks websites or network endpoints from test locations.