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 verification-skillgit 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/verification-skill)<a href="https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/verification-skill"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/verification-skill.svg" alt="Measured on agentmods" 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 | $0.00066 | $0.01384 |
| Opus 5 | $0.00033 | $0.00692 |
| Sonnet 5 | $0.00013 | $0.00277 |
| Haiku 4.5 | $0.00007 | $0.00138 |
Grade A, and why
verification-skill 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 3d 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 — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verification skill
A verification skill is a project-local, self-contained artifact that states one observable contract, the command that exercises it, and the pass condition. It executes without this skill.
Creating one and repairing one are two modes of the same artifact's lifecycle. Both write only inside the verification-skill directory, and both hold the same rule: when the harness disagrees with the repository, the harness is wrong.
Contract
| Field | Bound contract |
|---|---|
| Trigger | Create a project-local executable verification skill, or repair drift in an existing one |
| Authority | Reversible local writes inside the named verification-skill directory only. Never the source under test, CI config, credentials, remote state, or a deployed target |
| Side effect | Writes and runs a verification harness inside the project. Opens at most one pull request, and only on explicit approval |
| Done | Create: the harness runs and passes against the live repository. Repair: one honest classification of clean, changed, or blocked |
Mode selection
Create when no verification skill exists for the target contract. Repair when one exists and its commands, paths, or assertions no longer match the project. A repair that would have to rewrite the contract itself is a create, so say so and start over rather than editing around the mismatch.
Inputs
Create needs the feature, behavior, or contract to verify, and the verification target: a file, module, CLI command, or API surface. It discovers the target by repository interview when one is not supplied. Repair needs the verification-skill directory and the concrete drift evidence, which is the mismatch between the skill and current project state: a failing check, a stale command, an outdated path, or a changed contract. Both default the project root to the working directory. A pull request requires explicit approval; without it, edits stay local.
Procedure, create
- Interview the repository. Read the relevant source, tests, build config, and run commands to determine how the target is built, invoked, and observed. Done when: the build, run, and observe mechanism is determined from the repository rather than assumed.
- Define the verification contract. State the observable behavior, the inputs that exercise it, and the pass condition as concrete assertions. Done when: all three are stated.
- Bound the scope. Name the exact files to write, which is the skill directory and any harness script under it. Source under test, CI config, and unrelated paths are excluded. Done when: the write paths are named and the exclusions hold.
- Write the skill so it restates the contract, the run command, and the pass condition and executes on its own. Done when: the skill is written and self-contained.
- Run the harness live against the repository and capture the actual output. Done when: real output is captured, never predicted.
- Confirm every assertion holds against that output. When one fails, correct the harness or the run command and re-run. Never edit the source under test to make an assertion pass. Done when: every assertion holds, or the failing assertion is named for correction.
- Record the proof inside the skill: the exact command, its output, and the pass result. Done when: all three are recorded.
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.
- 3d ago First seen · 103 lines · 66 tokens per session scan A 2454fae21183
verification-skill is a skill published in the GitHub repository OutlineDriven/odin-claude-plugin (35 stars, last pushed today), licensed Apache-2.0. It adds 66 tokens to every session and 1,384 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 2026-09-04.
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…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…