Borrowing it
Nothing to install: this file belongs to dhruv-corethink/reasoning-traces. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/dhruv-corethink/reasoning-traces/main/CLAUDE.mdgit clone --depth 1 https://github.com/dhruv-corethink/reasoning-tracesWrote 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/instructions/dhruv-corethink/reasoning-traces/claude-md)<a href="https://agentmods.dev/instructions/dhruv-corethink/reasoning-traces/claude-md"><img src="https://agentmods.dev/badge/instructions/dhruv-corethink/reasoning-traces/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/dhruv-corethink/reasoning-traces/claude-md"><img src="https://agentmods.dev/badge/instructions/dhruv-corethink/reasoning-traces/claude-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00227 | $0.00227 |
| Opus 5.5 | $0.00091 | $0.00091 |
| Sonnet 5.5 | $0.00045 | $0.00045 |
| Haiku 4.5 | $0.00023 | $0.00023 |
Grade A, and why
reasoning-traces 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 2mo 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.
What it actually says
Reasoning Traces (development repo)
This repo is both the source and the Claude Code plugin for Reasoning Traces — an MCP server that consults a stronger reasoning model and returns its reasoning trace.
Layout: reasoning_traces/ is the Python package (FastMCP server + pluggable backends); .claude-plugin/ + mcp-config.json + commands/ make it an installable plugin; .mcp.json runs the server from source for dev sessions in this repo.
Using the oracle
- For any task involving multi-step reasoning — debugging subtle behavior, architecture decisions, algorithm design, tricky math — call
mcp__reasoning-traces__deep_reasoningBEFORE proposing a solution. Pass the problem plus all relevant code, errors, and context you have gathered; the reasoning model cannot see this conversation. - Treat the returned trace as advisory input: verify it against the actual code, then produce your own answer. Note where you disagree with it.
- For quick lookups, syntax questions, or mechanical edits, answer directly without the tool.
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.
- 2mo ago First seen · 12 lines · 227 tokens per session scan A 879b8f997d1b
reasoning-traces CLAUDE.md is an instructions file published in the GitHub repository dhruv-corethink/reasoning-traces (0 stars, last pushed 4mo ago), licensed MIT. It adds 227 tokens to every session, about $0.0009 per session on Opus 5.5. 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.
reasoning-traces CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| reasoning-traces CLAUDE.mddhruv-corethink/reasoning-traces | Instructions file | 0 | 4mo ago | MIT | A | 227 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |