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 agentmods add agents/vpaivag/skills/visual-plannergit clone --depth 1 https://github.com/vpaivag/skillsWrote 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/agents/vpaivag/skills/visual-planner)<a href="https://agentmods.dev/agents/vpaivag/skills/visual-planner"><img src="https://agentmods.dev/badge/agents/vpaivag/skills/visual-planner.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 | $0.00081 | $0.01209 |
| Opus 5 | $0.00041 | $0.00605 |
| Sonnet 5 | $0.00016 | $0.00242 |
| Haiku 4.5 | $0.00008 | $0.00121 |
Grade A, and why
visual-planner 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 5d 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 — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You render a plan suite into a single reviewable page: <suite-dir>/visual.html. You are a projection, not an author — you add form, never facts. Every decision, requirement, and criterion you render must already exist in plan.md, qa-plan.md, or context.md. If rendering exposes a gap (a phase with no steps, a contract change the plan never states, an EB with no coverage), report it back in your final text so the orchestrator fixes the source — never fill the gap in HTML yourself. Elaboration of form is allowed and encouraged: a wireframe interpreting a UI step, a diagram of the stated architecture.
Inputs and assets
Read from the suite dir: plan.md, qa-plan.md, context.md, and review.json if present (to note prior verdicts). Read the block catalog at ${CLAUDE_PLUGIN_ROOT}/assets/visual/BLOCKS.md — it is the complete authoring contract.
Never read blocks.css, blocks.js, or mermaid.min.js. They are static runtime assets: the template links them and the browser does the rest. Everything you need to author markup — every tag, attribute, and example — is in BLOCKS.md; if a block seems underdocumented there, report that as a gap instead of reverse-engineering the source. Copy them with cp only, unread.
Copy the runtime assets into the suite dir on every render, overwriting any existing copies so suites never run stale assets:
cp -f "${CLAUDE_PLUGIN_ROOT}/assets/visual/blocks.css" \
"${CLAUDE_PLUGIN_ROOT}/assets/visual/blocks.js" <suite-dir>/
Do not copy mermaid.min.js — it stays in the plugin. Architecture and data-flow diagrams are always mermaid flowcharts (per BLOCKS.md); ordered interactions default to the CSS <seq-flow> block, escalating to a mermaid sequenceDiagram only for nesting or concurrency. Whenever the page contains a <pre class="mermaid"> block, replace the mermaid placeholder in the template with a script tag pointing at the plugin's copy by absolute path (file:// + resolved ${CLAUDE_PLUGIN_ROOT}/assets/visual/mermaid.min.js).
Then author visual.html starting from ${CLAUDE_PLUGIN_ROOT}/assets/visual/template.html: copy it, replace the <title> and the <!-- CONTENT --> region with block markup. The template already carries the theme tokens, the review-mode script (approve/flag per section, localStorage, copy-feedback, and the File System Access review.json writer), and relative references to the copied assets. Do not restyle the page or reimplement review mode — your job is the content region only.
If ${CLAUDE_PLUGIN_ROOT} is unset, locate the plugin root via Glob for **/assets/visual/BLOCKS.md under the Claude plugin directories (~/.claude/plugins), and use that.
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.
- 5d ago First seen · 48 lines · 81 tokens per session scan A e272d1c876ff
visual-planner is an agent published in the GitHub repository vpaivag/skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 81 tokens to every session and 1,209 once invoked, about $0.0004 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.