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 googlicius/obsidian-steward --skill stateful-widgetgit clone --depth 1 https://github.com/googlicius/obsidian-stewardWrote 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/googlicius/obsidian-steward/stateful-widget)<a href="https://agentmods.dev/skills/googlicius/obsidian-steward/stateful-widget"><img src="https://agentmods.dev/badge/skills/googlicius/obsidian-steward/stateful-widget/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/googlicius/obsidian-steward/stateful-widget"><img src="https://agentmods.dev/badge/skills/googlicius/obsidian-steward/stateful-widget.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 9 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Prompt Injection · line 40 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Memory Poisoning · line 162 Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.Fix: Implement context-window management that detects and rejects padding or stuffing attempts. Prioritize system instructions over user-injected content.
- medium Memory Poisoning · line 164 Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.Fix: Implement context-window management that detects and rejects padding or stuffing attempts. Prioritize system instructions over user-injected content.
- medium Prompt Injection · line 162 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 164 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 165 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 166 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 167 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 168 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
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 | $0.00033 | $0.02712 |
| Opus 5 | $0.00016 | $0.01356 |
| Sonnet 5 | $0.00007 | $0.00542 |
| Haiku 4.5 | $0.00003 | $0.00271 |
Grade A, and why
stateful-widget 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 10d 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 — 209 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Stateful Widget Skill
Use this skill when creating interactive HTML project widgets (games, counters, quizzes, forms) where user actions must persist after the user closes and reopens the conversation note.
When to use
- User asks for a game, interactive demo, counter, quiz, or any widget that changes based on clicks/keyboard input.
- You use
show_widgetwith project mode (files: [{ name, content }, ...]), not a singlecodeblob. - State must survive: reopening the note, scrolling away and back, or editing widget code files (hot-reload).
Skip this pattern for static animations, one-shot diagrams, or SVG-only widgets with no runtime state.
How a widget is rendered
-
Create (
show_widget): Writes project files under{stewardFolder}/Widgets/{widgetId}/(index.html,main.js,style.css,Widget.md, etc.).widgetId= slugifiedwidgetName+ short unique suffix. Appends astw-widget-projectfence to the conversation note:<widgetId> -
Mount (
WidgetPostProcessor): When the note is displayed, the post-processor reads the fence, resolves{stewardFolder}/Widgets/{widgetId}, bundles the entry HTML (inlines linked CSS/JS from the project folder), injectswindow.stwfor state and assets, and renders the bundle in a sandboxed iframe (srcdoc,allow-scripts, strict CSP, no network). Manifestassetsare not inlined into HTML; the iframe requests allowed files from the host at runtime and creates local blob URLs. -
Hot-reload: Edits to project files via
edittrigger a re-bundle and iframe refresh.
Only project (HTML multi-file) widgets get the state bridge and state.json. Single-blob code widgets do not persist runtime state this way.
How state is managed
| Piece | Role |
|---|---|
window.stw |
API injected into the iframe by the host (not written by you in vault files). |
window.stw.getState() |
Returns last saved data object, or null on first load. |
window.stw.setState(data) |
Saves a JSON-serializable snapshot; debounced ~400ms, then written to vault. |
state.json |
Created lazily in the project folder on first successful setState. Host-owned; do not author or edit it manually. |
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.
- 10d ago First seen · 209 lines · 33 tokens per session scan A 07657ae3bc27
stateful-widget is a skill published in the GitHub repository googlicius/obsidian-steward (86 stars, last pushed 1mo ago), licensed MIT. It adds 33 tokens to every session and 2,712 once invoked, about $0.0002 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
spatie-guidelines
Spatie's PHP, Laravel, JavaScript and Vue coding guidelines and conventions. Use when writing or reviewing PHP, Laravel, JavaScript, or Vue code for Spatie projects or packages. Covers code style, type declarations, docblocks, control flow, naming, routing, controllers, Blade, validation, comments, testing (Pest)…
laravel-inertia-react-structure
Frontend structure conventions for Laravel Inertia React applications based on Spatie's production practices. Use when creating, scaffolding, or reviewing frontend code in a Laravel Inertia React project. Triggers on creating React components, pages, modules, organizing frontend directories, setting up Inertia pages…
react-best-practices
Provides React patterns for hooks, effects, refs, and component design. Covers escape hatches, anti-patterns, and correct effect usage. Must use when reading or writing React components (.tsx, .jsx files with React imports).
livewire-4
Build Livewire 4 components and applications. Use when creating reactive Laravel interfaces, single-file components, multi-file components, page components, forms, actions, or any Livewire-based UI. Triggers include "Livewire component", "wire:model", "wire:click", building interactive Laravel UIs, or working with…
worker-visualizer
A real-time data/particle/simulation visualizer whose heavy compute runs in a Web Worker (off the main thread), optionally sharing memory with the UI via SharedArrayBuffer, and renders to a canvas at 60fps. Produced as a single self-contained index.html. Use when the brief asks for a "web worker", "simulation"…
develop-web-game
Use when Codex is building or iterating on a web game (HTML/JS) and needs a reliable development + testing loop: implement small changes, run a Playwright-based test script with short input bursts and intentional pauses, inspect screenshots/text, and review console errors with rendergametotext.