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 vinsonconsulting/claude-skill-foundry --skill textualgit clone --depth 1 https://github.com/vinsonconsulting/claude-skill-foundryWrote 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/vinsonconsulting/claude-skill-foundry/textual)<a href="https://agentmods.dev/skills/vinsonconsulting/claude-skill-foundry/textual"><img src="https://agentmods.dev/badge/skills/vinsonconsulting/claude-skill-foundry/textual/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/vinsonconsulting/claude-skill-foundry/textual"><img src="https://agentmods.dev/badge/skills/vinsonconsulting/claude-skill-foundry/textual.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 | $0.00294 | $0.03966 |
| Opus 5 | $0.00147 | $0.01983 |
| Sonnet 5 | $0.00059 | $0.00793 |
| Haiku 4.5 | $0.00029 | $0.00397 |
Grade A, and why
textual 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.
How it starts
The opening of the file, as written. The whole thing — 318 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Textual
Write current, compiling Textual code (pinned to 8.2.7, Python ≥3.9, Rich 15.0.0)
and refuse the pre-1.0 / 0.x patterns the model remembers from training — the 1.0 (Dec 2024)
and 2.0 (Feb 2025) releases were hard breaks. The body is the load-bearing 20%: one mental
model and one run-verified example per concept. Everything enumerable — the full widget
catalog, every TCSS rule, the streaming model, testing, deploy, migration — lives in
references/. Open the matching reference before writing nontrivial code in that area.
Mental model
Textual is a retained, reactive, DOM-like framework — the opposite of Ratatui's immediate mode and Bubble Tea's MVU. You build a tree of widget objects once, then mutate their state; Textual re-renders only the affected parts, like a web framework. It is async-native (asyncio) and styled with Textual CSS (TCSS), not layout math.
Four nouns carry everything:
- App — the application and event loop;
App().run()(orawait run_async()). Holds screens, handles input, owns the@workworkers. - Screen — a full-window container you push/pop; the default screen hosts your
compose(). Modals/dialogs are screens. - Widget — a node in the DOM tree. Leaf widgets draw themselves (
render()); compound widgets yield children (compose()). - DOM + TCSS — widgets form a tree you query with
query_one/query(CSS selectors) and style with TCSS. Mutating a widget'sreactivestate schedules a repaint.
App, compose, and lifecycle
compose() runs once to build the tree; never touch widgets there — the DOM isn't mounted
yet. Wait for on_mount, then resolve widgets by selector with query_one.
from textual.app import App, ComposeResult
from textual.widgets import Header, Footer, Button, Label
class CounterApp(App):
BINDINGS = [("a", "add", "Add"), ("q", "quit", "Quit")] # key → action_* → footer hint
def compose(self) -> ComposeResult: # build the tree ONCE
yield Header()
yield Label("count: 0", id="lbl")
yield Button("hit", id="btn")
yield Footer()
def on_mount(self) -> None: # DOM is live; safe to touch widgets
self.count = 0
def action_add(self) -> None: # bound to "a"
self.count += 1
self.query_one("#lbl", Label).update(f"count: {self.count}")
if __name__ == "__main__":
CounterApp().run()
What ships with it
17 files 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.
- card-review.md 868 B
- card.authored.yaml 450 B
- card.json 4.2 KB
- evals/evals.json 5.2 KB
- README.md 4.5 KB
- references/agent-ui.md 4.1 KB
- references/architecture.md 6.7 KB
- references/ecosystem.md 2.3 KB
- references/styling-and-layout.md 4.2 KB
- references/testing.md 2.9 KB
- references/text-and-unicode.md 2.4 KB
- references/versioning.md 3.8 KB
- references/web-deploy.md 2.3 KB
- references/widgets.md 4.2 KB
- scan.json 2.2 KB
- scripts/verify.py 11 KB runs code
- skill-card.md 4.9 KB
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 · 318 lines · 294 tokens per session scan A 480215aad8a1
textual is a skill published in the GitHub repository vinsonconsulting/claude-skill-foundry (1 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 294 tokens to every session and 3,966 once invoked, about $0.0015 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 2026-08-31.
Other skills, from other repositories
streamlit
Streamlit Python web application framework. Covers session state, caching, layouts, widgets, multipage apps, and deployment. Use when building interactive Python data apps or dashboards. USE WHEN: user mentions "streamlit", "st.sessionstate", "st.cachedata", "streamlit app", "python dashboard", "python web app"…
accelerate
Run PyTorch training across GPUs with minimal changes.
jupyter-notebook
Iterative Python via live Jupyter kernel (hamelnb).
temporal-python-testing
Test Temporal workflows with pytest, time-skipping, and mocking strategies. Covers unit testing, integration testing, replay testing, and local development setup. Use when implementing Temporal workflow tests or debugging test failures.
huggingface-gradio
Build Gradio web UIs and demos in Python. Use when creating or editing Gradio apps, components, event listeners, layouts, or chatbots.
shiny-for-python
Building, styling, testing, debugging, or observing a Shiny for Python (py-shiny) reactive web app - from shiny import ..., shiny run app.py. Index skill: read this, then open the linked reference for the task. Covers dashboard design and visual QA; card toolbars and accessible icons; interactive Plotly charts and…