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 instructions/viewflow/seedkit/claude-mdgit clone --depth 1 https://github.com/viewflow/seedkitWhat 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 | $0.01325 | $0.01325 |
| Opus 5 | $0.00662 | $0.00662 |
| Sonnet 5 | $0.00265 | $0.00265 |
| Haiku 4.5 | $0.00133 | $0.00133 |
Grade A, and why
seedkit CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
seedkit
Claude Code plugin: a skill for bootstrapping Django projects. The repo root is the plugin root, so everything here ships to anyone who installs it.
Layout:
.claude-plugin/plugin.json— the plugin manifest..claude-plugin/marketplace.json— this repo serving as its own single-plugin marketplace.skills/django-seedkit/SKILL.md+skills/django-seedkit/references/*.md— the skill itself.testcases/0[1-9]-*.md— scripted runs that exercise the skill end-to-end. Specs, not scripts: the harness that runs them lives elsewhere.
Two sibling repos hold everything that is not the skill. Keep it that way — a plugin repo is downloaded in full by every user, and its whole tree is read by the marketplace review pipeline.
- seedkit-examples — the generated projects and the
train/harness that produces them. It readsskills/andtestcases/from here via$SEEDKIT(default../seedkit) and is the only thing that writes back. ItsCLAUDE.mdholds the harness and baseline contracts. - viewflow-pro/seedkit/ (private) — the django-seedkit.viewflow.io landing page and blog, its
deploy playbook, and the article backlog. Its
CLAUDE.mdholds the blog rules.
Run claude plugin validate . after touching either manifest.
Writing reference files
Each reference is a paste-ready snippet plus the minimum prose needed to use it correctly.
Show the correct sample. Imperative, specific, copyable. The agent uses snippets verbatim (SKILL.md "Snippet integrity" pitfall) — anything not in the snippet won't make it into the generated project.
Drop the matching "don't". When a positive sample shows the right way, don't follow it with "Don't write X" / "Never use Y". The correct line is the canonical answer; the redundant warning adds tokens, invites the agent to second-guess, and ages badly. Keep negative guidance only when it stands alone — a behavior rule with no positive code sample (e.g. "Don't strip the Host-header check globally", "Don't log request bodies").
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.
- 2d ago First seen · 82 lines · 1,325 tokens per session scan A 0467fba0a298
seedkit CLAUDE.md is an instructions file published in the GitHub repository viewflow/seedkit (54 stars, last pushed 10d ago), licensed MIT. It adds 1,325 tokens to every session, about $0.0066 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 instructions, from other repositories
claude-ads CLAUDE.md
Instructions for AgriciDaniel/claude-ads, covering claude ads repository instructions, architecture, development and verification.
notebooklm-skill AGENTS.md
Instructions for claude-world/notebooklm-skill, covering notebooklm-skill, authentication, cli commands, notebooklm-skill — core operations and notebooklm-pipeline — workflow orchestration.
recursive-decomposition-skill AGENTS.md
Instructions for massimodeluisa/recursive-decomposition-skill, covering agents: recursive-decomposition skill, read these first (mandatory), language policy, non-negotiables and commands.
stockbit-mcp CLAUDE.md
Instructions for INo-xious/stockbit-mcp, covering claude.md, what this is, commands, the map and three invariants. do not break them; each has a test.
devin-handoff AGENTS.md
Instructions for club-cog/devin-handoff, covering devin handoff — agent guide, what this is, when to hand off, finding the script and how to use.
agent-plugins-skills copilot-instructions.md
Instructions for richfrem/agent-plugins-skills, covering copilot instructions for agent-plugins-skills, 1. think before coding, 2. simplicity first, 3. surgical changes and 4. goal-driven execution.