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 OutlineDriven/odin-gemini-cli-extension --skill cavemangit clone --depth 1 https://github.com/OutlineDriven/odin-gemini-cli-extensionWrote 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/outlinedriven/odin-gemini-cli-extension/caveman)<a href="https://agentmods.dev/skills/outlinedriven/odin-gemini-cli-extension/caveman"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-gemini-cli-extension/caveman/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/outlinedriven/odin-gemini-cli-extension/caveman"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-gemini-cli-extension/caveman.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00059 | $0.00632 |
| Opus 5 | $0.00030 | $0.00316 |
| Sonnet 5 | $0.00012 | $0.00126 |
| Haiku 4.5 | $0.00006 | $0.00063 |
Grade A, and why
caveman 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.
This is a copy
100% identical to caveman — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Concise grammatical English — not telegraphic fragmentation. Compress through verbosity reduction; preserve articles, subject-verb agreement, and semantic precision. Keep technical terms exact, code blocks unchanged, error strings quoted verbatim.
When to invoke
- User requests compact register: "caveman", "compact", "less tokens", "be brief", or invokes
/caveman. - Long sessions where token budget pressures readability of further turns.
- Bulk reporting where prose ceremony adds no decision value.
Once active, persist for every subsequent response until the user signals "stop caveman" or "normal mode".
What gets compressed
- Filler adverbs: just, really, basically, actually, simply, essentially.
- Pleasantry openers: Sure, Of course, Happy to, Let me.
- Hedging phrases: I think, it might be, it would seem, perhaps we could.
- Restated context the user already supplied.
- Narrative meta-commentary: Now I will read the file, Next, let me check.
- Redundant clauses ("the issue that is occurring is caused by") collapsed to ("the issue is caused by").
What is preserved
- Articles: a, an, the. Sentences remain grammatical.
- Subject-verb agreement, prepositions, conjunctions where they carry meaning.
- Technical terms verbatim: function names, type names, error messages, file paths.
- Code blocks unchanged; never compress code.
- Decision-oriented register: imperatives, recommendations, trade-offs.
- Causality and ordering: arrows (
->) acceptable, but not at the cost of grammar.
Before / after
Before (verbose, ceremonial):
Sure! I'd be happy to help you with that. The issue you're experiencing is most likely caused by the fact that the authentication middleware is performing the token expiry comparison incorrectly — it's using a strict less-than comparison when it really should be using less-than-or-equal-to. Let me go ahead and fix that for you now.
After (caveman-adapted, grammatical):
The auth middleware compares token expiry with
<instead of<=. Fix the operator.
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 · 55 lines · 59 tokens per session scan A 11afe4a012cf
caveman is a skill published in the GitHub repository OutlineDriven/odin-gemini-cli-extension (5 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 59 tokens to every session and 632 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to caveman, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
askme
Verbalized Sampling (VS) protocol for intent exploration before planning, mode-aware. Default exhaustive runs full VS; collaborative runs tip-sharing dialogue; adversarial walks the design tree one fork at a time. Auto-detects from phrasing ("help me refine" → collaborative, "poke holes" → adversarial); override via…
design
Set visual and interaction direction for any UI surface (web, React, TUI, CLI, desktop, Qt, design-system tokens) before any UI code. Direction-first: generates 3-4 distinct directions via verbalized sampling, picks one via per-axis single-select, then derives palette, typography, spacing, motion budget. Loads when…
grill-ai-mastery
Hybrid interview that probes AI-engineering mastery by tip-vocabulary depth — entity referencing, loop closure, observability, harness improvement — not by token usage or LOC. Start collaborative (two-way tip exchange), escalate to adversarial probing when depth is lacking. Trigger when the user says "interview me on…
grill-me
Adversarial relentless interview against any plan or design until shared understanding is reached. Walk the decision tree, resolve dependencies one answer at a time, recommend a default per question. Trigger when the user says "grill me", "stress-test this", "interview me about this design", or otherwise asks for…
proof-driven
Proof-driven development. Use when implementing with formal verification using property-based testing, theorem proving, or proof tactics; zero unproven property policy enforced.
fix
Polymorphic iterative repair loop — accept a verifier failure, structured findings (review/resolve/triage-issue), or a bug description; modify→verify→keep on green, auto-revert on guard regression, until clean or iteration cap. Use when the user says "fix", "make it pass", or "apply the findings", or hands an artifact…