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 possible-worlds-qagit 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/possible-worlds-qa)<a href="https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/possible-worlds-qa"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/possible-worlds-qa/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-claude-plugin/possible-worlds-qa"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/possible-worlds-qa.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00086 | $0.01496 |
| Opus 5.5 | $0.00034 | $0.00598 |
| Sonnet 5.5 | $0.00017 | $0.00299 |
| Haiku 4.5 | $0.00009 | $0.00150 |
Grade A, and why
possible-worlds-qa 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 1mo 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Possible worlds QA
A passing test is not evidence that a surface is sound; it is evidence that one world was survivable. This skill enumerates the worlds a surface can find itself in, then either measures what the surface fails to cover or proves it can be broken.
Contract
| Field | Bound contract |
|---|---|
| Trigger | A named product surface must be tested against extreme or hostile worlds rather than a green-check suite. |
| Authority | Write only named local test cases and report artifacts, and state the rollback path before the first write. No source, version control, credential, paid, published, deployed, or remote mutation. |
| Side effect | Test-case files and one report written to the local working tree. Deleting them reverses the run. |
| Done (completeness) | Every enumerated world is classified covered, gap, or untestable against a stated criterion, and the report names the world, observed behaviour, violated criterion or blocker, and test-case path for each. |
| Done (break) | Every enumerated world is broken with an attached artifact, escalated until exhausted, or recorded untestable with its blocker, and the report carries the world catalogue, break log, survived log, and summary. |
Mode selection
| Mode | Select when | Verdict it produces |
|---|---|---|
| Completeness | The caller wants to know what the surface fails to handle, measured against criteria it claims to meet. | Each world is covered, a gap, or untestable. |
| Break | The caller wants proof the surface can be made to fail. | Each world breaks the surface, survives and is escalated until exhausted, or is untestable with a named blocker. |
Ask when the request names neither. The modes share steps 1 through 4 and diverge at step 5.
Inputs
- The product surface, required: a named code path, module, API, binary, feature, or system boundary reachable from the local environment.
- Completeness criteria, required in completeness mode: the invariants, contracts, or acceptable behaviours the surface claims. When absent, derive minimal criteria from the surface's own stated invariants; when the surface states none, report completeness undefined and stop.
- World budget, optional: a cap on enumerated worlds. When absent, enumerate until every distinct dimension the surface exposes is represented once.
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.
- 1mo ago First seen · 61 lines · 86 tokens per session scan A dad0e80bc203
possible-worlds-qa is a skill published in the GitHub repository OutlineDriven/odin-claude-plugin (39 stars, last pushed 26d ago), licensed Apache-2.0. It adds 86 tokens to every session and 1,496 once invoked, about $0.0003 per session on Opus 5.5. 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
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…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
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…
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.
component-fixtures
Use when creating or updating component fixtures for screenshot testing or their shared infrastructure, or when designing UI components to be fixture-friendly. Covers fixture file structure, theming, service setup, CSS scoping, async rendering, validation, and common pitfalls.