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/corezoid/simulator-ai-plugin/claude-mdgit clone --depth 1 https://github.com/corezoid/simulator-ai-pluginWhat 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 | $0.00895 | $0.00895 |
| Opus 5 | $0.00447 | $0.00447 |
| Sonnet 5 | $0.00179 | $0.00179 |
| Haiku 4.5 | $0.00089 | $0.00089 |
Grade A, and why
simulator-ai-plugin 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 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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Guidance for Claude Code when working in the simulator-ai-plugin repository.
The canonical, tool-agnostic instructions live in
AGENTS.md— repo overview, layout, commands, conventions, and gotchas. Read it first. This file only adds Claude Code–specific notes; keep substantive guidance inAGENTS.mdto avoid drift.
Quick orientation
- What it is: an MCP plugin connecting the Simulator.Company platform to Claude — a Go
MCP server (
plugins/simulator/mcp-server/) exposing the REST API as tools, plus 6 skills (plugins/simulator/skills/). - Design:
docs/ARCHITECTURE.md. - Verify changes:
make build && make vetfrom the repo root (make lintis available — golangci-lint v2, advisory:gosecclean, style backlog open).
Claude Code–specific notes
- Skills. The
simulator*skills here are the ones surfaced as/simulator,/simulator-graph, etc. When editing a skill's behaviour, edit itsSKILL.md; if you change the frontmatter, regenerate discovery artifacts withmake discovery(don't hand-editpublic/). Write instructions in English and have Claude reply in the user's own language — never hardcode a non-English sentence for it to say. See the "Skill language" convention inAGENTS.mdfor the trilingual policy (en primary; uk/ru activation triggers and product aliases are the kept exceptions). $CLAUDE_PLUGIN_ROOT=plugins/simulator/. Skills load reference docs as$CLAUDE_PLUGIN_ROOT/docs/entities/*.md. Those files must stay underplugins/simulator/docs/(only the plugin dir is shipped on install). Contributor docs go in the repo-rootdocs/. Claude Code and Codex both resolve this exact token via text substitution at skill-load time; renaming it breaks doc loading on both hosts (anthropics/claude-code#48230, #47789, #44057). For AWS Kiro — which does no such substitution —install-kiro.shand the release-zip generator hard-copy the skills andsed-replace the token with the absolute plugin path at install time.- MCP tools. When you add or rename a tool, update the MCP-tools table in the root
README.mdand §4 ofdocs/ARCHITECTURE.md. - Curated tools live in Go under
internal/tools/<domain>.go(declared as typedOperations), not generated from a spec. The drift gate (internal/tools/testdata/papi-openapi.json) validates them against the backend. - Don't hand-edit generated files:
public/*(usemake discovery). Refresh the drift spec from pong-server (yarn dump-openapi), don't write it by hand.
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 · 55 lines · 895 tokens per session scan A e04fe5f6b2fb
simulator-ai-plugin CLAUDE.md is an instructions file published in the GitHub repository corezoid/simulator-ai-plugin (60 stars, last pushed 6d ago), licensed MIT. It adds 895 tokens to every session, about $0.0045 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
unslop CLAUDE.md
Instructions for MohamedAbdallah-14/unslop, covering unslop — maintainer guide, readme is a product artifact, what this repo is, source of truth and mirrored locations (do not edit by hand).
unslop AGENTS.md
Instructions for MohamedAbdallah-14/unslop, covering agent instructions — unslop repo, workflow, tone, commits and reviews.
unslop copilot-instructions.md
Instructions for MohamedAbdallah-14/unslop: When generating chat replies, code comments, or commit messages in this repository.
unslop GEMINI.md
Instructions for MohamedAbdallah-14/unslop, covering unslop — gemini context, drop, keep, pattern and boundaries.
courier CLAUDE.md
Instructions for yegor256/courier: Write one sentence per line. Keep each sentence short and concise. Separate clauses with periods rather than semicolons, colons, or em dashes. Name action, target, and condition. Write rules as direct orders, not opinions or suggestions. Strip emotions and rhetorical flourish. Forbid…
Adversaria CLAUDE.md
Claude Code instructions for damionrashford/Adversaria, covering claude.md, what this repository is, development tooling, commands and architecture.