Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/ChenZiHong-Gavin/llm-tech-reportnpx agentmods add rules/chenzihong-gavin/llm-tech-report/logo-generationWrote 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/rules/chenzihong-gavin/llm-tech-report/logo-generation)<a href="https://agentmods.dev/rules/chenzihong-gavin/llm-tech-report/logo-generation"><img src="https://agentmods.dev/badge/rules/chenzihong-gavin/llm-tech-report/logo-generation.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 5.1 | $0.00048 | $0.00647 |
| Opus 5 | $0.00024 | $0.00324 |
| Sonnet 5 | $0.00010 | $0.00129 |
| Haiku 4.5 | $0.00005 | $0.00065 |
Grade A, and why
logo-generation 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 7d 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 — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Logo Generation (Timeline)
Generates the timeline logo.png for this repo. All model names are stacked above and below a horizontal time axis, with automatic overlap avoidance.
File layout
.claude/skills/logo-generation/
├── SKILL.md # canonical doc
└── scripts/
├── generate.py # entry point: config + invoke
├── models.py # data: all (year, name, company) entries
└── draw.py # drawing: timeline + anti-overlap algorithm
Output: logo.png in the repo root (~400 KB, 20×8 in, 300 DPI).
Dependencies
pip install matplotlib numpy
Run
# from the repo root
python .claude/skills/logo-generation/scripts/generate.py
How to update
Add a model
Edit scripts/models.py, add to the MODELS list:
(2025.50, "ModelName", "CompanyKey"),
- year:
YYYY + MM/12(e.g. 2025-06 →2025.50) - name: keep it short (
DS-R1, notDeepSeek-R1) - company: must already exist in the
COLORSdict ingenerate.py
No manual sorting needed — the script sorts by time.
Add a company
- Add
"NewCo": "#hex"to theCOLORSdict ingenerate.py - Add the name to the
legendlist indraw.pyif it should show in the legend
Adjust landmarks
Add/remove names in the LANDMARKS set in generate.py. Landmark font is 7.5 pt, normal 5.8 pt.
Anti-overlap algorithm
- Pure data-coordinate estimation (no renderer call) — runs in seconds
- text width =
chars × fontsize × 0.55 / 72 × (x_range / fig_w) - text height =
fontsize × 1.2 / 72 × (y_range / fig_h) - alternates above/below the axis by parity; on collision, pushes outward (step 0.24), up to 30 layers
Notes
- Keep names short — long names crowd the axis; prefer abbreviations
- 2023–2025 is the densest stretch — nudge the year decimal to de-cluster
- Past ~150 models — increase
FIG_Hor decreasey_step - Dark background — avoid very dark colors (pure black, deep gray)
- Re-run after any data change and eyeball the result for overflow
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.
- 7d ago First seen · 68 lines · 48 tokens per session scan A 16e455191b9c
logo-generation is a cursor rule published in the GitHub repository ChenZiHong-Gavin/llm-tech-report (35 stars, last pushed 18d ago), licensed MIT. It adds 48 tokens to every session and 647 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 cursor rules, from other repositories
080-creative-phase-metrics
description: WHEN entering a creative phase ENSURE comprehensive exploration and validation globs: ["/.md", "/.mdc"] alwaysApply: true.
livkiet-plugins-playai
Agent Framework plugin for voice synthesis with PlayAI API.
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
cli-error-handling
CLI command error handling patterns.