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-claude-plugin --skill ruzzygit clone --depth 1 https://github.com/OutlineDriven/odin-claude-pluginWrote 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-claude-plugin/ruzzy)<a href="https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/ruzzy"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/ruzzy.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 | $0.00051 | $0.01437 |
| Opus 5 | $0.00026 | $0.00718 |
| Sonnet 5 | $0.00010 | $0.00287 |
| Haiku 4.5 | $0.00005 | $0.00144 |
Grade A, and why
ruzzy 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 yesterday.
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 — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ruzzy fuzzing campaign
Contract
| Field | Bound contract |
|---|---|
| Trigger | User needs Ruzzy to run coverage-guided fuzzing on Ruby code or a Ruby native extension. |
| Authority | Reversible local: writes harness, tracer, sanitizer preload, corpus, and installed gem files to the working directory or the local Ruby gem path. Record the installed-gem list before install; rollback is file deletion plus uninstalling exactly the gem version this campaign added. No remote mutation. |
| Side effect | Write Ruzzy harness scripts, tracer scripts, sanitizer LD_PRELOAD paths, corpus files, and install the target gem with clang and sanitizer flags. |
| Done | Ruzzy executes the intended Ruby target with the correct tracer or extension setup and reproduces saved failures. |
Inputs
Required: Ruby target path or gem name, sanitizer selection (ASan or UBSan).
Optional: corpus directory path, libFuzzer arguments (e.g., -max_len=1024), crash file to reproduce.
Procedure
- Confirm target: pure Ruby (requires tracer) or C extension (single harness). Done when: the target kind is confirmed as pure Ruby or C extension.
- Confirm sanitizer: ASan (
Ruzzy::ASAN_PATH) or UBSan (Ruzzy::UBSAN_PATH). Done when: the sanitizer is selected and the selectedRuzzy::ASAN_PATHorRuzzy::UBSAN_PATHconstant is confirmed. - For pure Ruby targets, write a tracer script calling
Ruzzy.trace('harness.rb')and a separate harness script callingRuzzy.fuzz(test_one_input). For C extensions, write one harness script callingRuzzy.fuzz(test_one_input); no tracer required. Done when: the harness and, for pure Ruby, the tracer script are written. - Write the harness as a lambda named
test_one_inputthat accepts data and returns0. Catch Ruby exceptions in C extension harnesses; let them propagate in pure Ruby harnesses. Done when: the harness lambda is written with the correct exception handling for the target kind. - Set
ASAN_OPTIONS=allocator_may_return_null=1:detect_leaks=0:use_sigaltstack=0. Do not exportLD_PRELOAD; use it inline with the ruby command. Done when:ASAN_OPTIONSis set andLD_PRELOADis prepared for inline use. - Record
gem list --local(each gem's name and version) before any install. Then install the gem with clang and sanitizer flags: setCC=clang,CXX=clang++,LDSHARED="clang -shared", andLDSHAREDXX="clang++ -shared"; setCFLAGSandCXXFLAGSto-fsanitize=address,fuzzer-no-link -fno-omit-frame-pointer -fno-common -fPIC -gfor an ASan build, or-fsanitize=undefined,fuzzer-no-link -fno-omit-frame-pointer -fno-common -fPIC -gfor a UBSan build, matching the sanitizer selected in step 2 so the instrumented library the run preloads matches the build. Also setMAKE="make --environment-overrides V=1"so GNU make lets environmentCC,CFLAGS, and related values override the mkmf Makefile; without--environment-overrides, Ruby's extension build can ignore those flags and install an uninstrumented native library. The-sharedflag is required forLDSHAREDandLDSHAREDXX; without it the native extension link fails. Done when: the pre-install gem list is recorded and the gem is installed with clang and the flags matching the selected sanitizer. - Run:
LD_PRELOAD=$(ruby -e 'require "ruzzy"; print Ruzzy::ASAN_PATH') ruby "<harness-or-tracer>.rb" [corpus] [libfuzzer-options]for ASan, or setSAN_PATH=$(ruby -e 'require "ruzzy"; print Ruzzy::UBSAN_PATH')and runLD_PRELOAD="$SAN_PATH" ruby "<harness-or-tracer>.rb" [corpus] [libfuzzer-options]for UBSan. Done when: the fuzzer runs with the correctLD_PRELOADand harness. - On
ERROR: AddressSanitizer:orERROR: UndefinedBehaviorSanitizer:, capture the crash file path, Base64 content, and reproducer command. Writecrash-*files to the working directory. Done when: any sanitizer error is captured with crash file path, content, and reproducer command, or the campaign completes without error. - To reproduce a saved failure, run the same command passing the crash file path as the final argument. Done when: the saved failure is reproduced or confirmed non-reproducible.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- yesterday First seen · 56 lines · 51 tokens per session scan A 16d227337502
ruzzy is a skill published in the GitHub repository OutlineDriven/odin-claude-plugin (35 stars, last pushed today), licensed Apache-2.0. It adds 51 tokens to every session and 1,437 once invoked, about $0.0003 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 2026-09-06.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…