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 skills/robinslange/learning-loop/diagramnpx skills add robinslange/learning-loop --skill diagramgit clone --depth 1 https://github.com/robinslange/learning-loopWrote 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/robinslange/learning-loop/diagram)<a href="https://agentmods.dev/skills/robinslange/learning-loop/diagram"><img src="https://agentmods.dev/badge/skills/robinslange/learning-loop/diagram.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 | $0.00054 | $0.01098 |
| Opus 5 | $0.00027 | $0.00549 |
| Sonnet 5 | $0.00011 | $0.00220 |
| Haiku 4.5 | $0.00005 | $0.00110 |
Grade A, and why
diagram 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 — 115 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Diagram: Excalidraw Generation
Overview
Generate an Excalidraw diagram and save it to the vault. Handles the full pipeline: determine what to diagram, build the JSON, write the file, and optionally embed it in a note.
When to Use
/learning-loop:diagram "concept or mechanism"-- user specifies what to diagram/learning-loop:diagram-- infer from conversation context
Process
Step 1: Determine What to Diagram
If args provided: Use as the concept description.
If no args: Read recent conversation. Identify the most diagram-worthy concept -- a mechanism, pathway, feedback loop, multi-step process, or system architecture where relationships between parts matter more than the parts themselves.
Diagram gate -- check BEFORE proceeding:
- Does the concept involve 3+ connected components?
- Do arrows/connections carry meaning (causation, flow, dependency)?
- Would prose obscure the relationships?
If the answer to all three is no, tell the user: "This works better as text/table -- no diagram needed." and stop.
Step 2: Load Diagram Rules
Read the diagram-rules agent file to get the full Excalidraw construction spec:
${CLAUDE_PLUGIN_ROOT}/agents-shared/diagram-rules.md
Follow every rule in that file. Key points:
.excalidraw.mdformat with empty## Text Elementssection- Soft fill colors:
#a5d8ff(blue),#b2f2bb(green),#ffec99(yellow),#ffc9c9(red),#d0bfff(purple) fontFamily: 5,roughness: 1- Descriptive string IDs, random seed integers
- Do NOT include
frameId,index,versionNonce, orrawTextfields - Text width: ~10px/char at 16, ~13px/char at 20, ~16px/char at 28
- Container width = text width + 60px padding
- 30px minimum gap between element edges
- 4-8 core elements. If more than 15, split or simplify.
Step 3: Plan the Layout
Before writing JSON, plan on paper:
- Identify elements: List the 4-8 core components
- Identify connections: What arrows connect what? Solid (direct) or dashed (indirect)?
- Choose pattern: Fan-out, convergence, timeline, cycle, assembly line, or diamond decision
- Choose layout direction: Left-to-right for sequences, top-to-bottom for hierarchies
- Calculate positions: Space elements 200px apart minimum, 30px gap between edges
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 · 115 lines · 54 tokens per session scan A 9cd509973dc4
diagram is a skill published in the GitHub repository robinslange/learning-loop (11 stars, last pushed 12d ago), licensed Apache-2.0. It adds 54 tokens to every session and 1,098 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 2026-08-30.
Other skills, from other repositories
ss-learn
Capture a human-approved UI design lesson as a privacy-minimized local StyleSeed candidate, review it, and prepare an opt-in share package without transmitting project code, prompts, screenshots, or brand data. Use when a person asks StyleSeed to remember, learn from, generalize, review, or prepare a reusable rule…
bmad-correct-course
Assess the impact of a significant change during sprint execution across the PRD, epics, architecture, and UX documents, and produce a sprint change proposal. Use when the user says "correct course" or "propose sprint change".
docs-workflows
Documentation and extension workflows derived from local slash commands (docs, explanation, issues, prototypes, tutorials, MCP design review).
Accessibility Manual Audit
Teach agents to guide manual accessibility audits for keyboard, screen reader, zoom, reflow, focus, and WCAG 2.2 criteria that scanners miss.
github-a11y-scanner
Integration patterns for the GitHub Accessibility Scanner Action (github/accessibility-scanner). Teaches agents how to detect scanner presence, parse scanner-created issues, correlate findings with local scans, and track Copilot-assigned fix status.
lighthouse-scanner
Integration patterns for Lighthouse CI accessibility auditing. Teaches agents how to detect Lighthouse CI configuration, parse accessibility audit results, map findings to the standard severity model, correlate with local axe-core scans, and track score regressions.