Borrowing it
Nothing to install: this file belongs to pguilp25/jarvis. 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/pguilp25/jarvis/main/CLAUDE.mdgit clone --depth 1 https://github.com/pguilp25/jarvisWrote 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/pguilp25/jarvis/claude-md)<a href="https://agentmods.dev/instructions/pguilp25/jarvis/claude-md"><img src="https://agentmods.dev/badge/instructions/pguilp25/jarvis/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/pguilp25/jarvis/claude-md"><img src="https://agentmods.dev/badge/instructions/pguilp25/jarvis/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 | $0.05730 | $0.05730 |
| Opus 5.5 | $0.02292 | $0.02292 |
| Sonnet 5.5 | $0.01146 | $0.01146 |
| Haiku 4.5 | $0.00573 | $0.00573 |
Grade A, and why
jarvis 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.
How it starts
The opening of the file, as written. The whole thing — 134 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
What JARVIS is
A multi-brain AI agent that coordinates many free/weak LLMs (NVIDIA NIM, OpenRouter, DeepInfra, Groq, Gemini) to approximate frontier-model quality at $0, slower. The bet: several models plan independently, critique, and merge, beating one weak model thinking alone. The hardest and most actively developed part is the coding agent — a 4-role pipeline (UNDERSTAND → PLAN → IMPLEMENT → REVIEW) whose target is to score well on SWE-bench / SWE-bench Pro with weak models.
Parallel-instance coordination (READ FIRST — two Claude Code instances are active)
Two Claude Code instances work this repo at the same time. Stay in your lane; flag anything that crosses lanes before you touch it. Assume the other instance has unrelated edits in flight — never git checkout/revert/stash broadly, never "clean up" files you don't own, and git pull/rebase mindset: your working tree is shared.
- Lane A — Coding agent (the UNDERSTAND→PLAN→IMPLEMENT→REVIEW pipeline + its SWE-bench eval). Files:
core/native_tools.py,core/tool_call.py,core/tool_detector.py,core/prompts_v8.py,core/self_verify.py,core/structural_gate.py,core/remote_check.py,core/review_verify.py,core/verify_toolkit.py,core/plan_*.py,core/code_index.py,core/cli.py(trace),workflows/code.py,tools/,swe_bench.py+behavioral_audit/(coding-agent eval/grading). - Lane B — Context system (a NEW subsystem giving the MAIN PIPELINE better context; explicitly NOT the coding agent — used by the main pipeline, not the coder/planner/reviewer). Files: the new context-system module(s) + the main-pipeline hooks that gather/inject context. (Lane B owner: list your files here.)
Shared / cross-cutting — touching these can impact the other lane → ANNOUNCE first (log line + commit note):
main.py, ui_main.py and the top-level pipeline orchestration; clients/ (the LLM clients both lanes call); shared config / env-key handling; CLAUDE.md; and — the one real seam — the handoff where the main pipeline passes the task/context into the coding agent. If Lane B changes what context the coder/planner actually receives, that DOES reach Lane A → flag it loudly (Lane A's whole doctrine is "verify the rendered artifact the model receives"; silently changing that artifact breaks it).
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 · 134 lines · 5,730 tokens per session scan A 2d976156e9dd
jarvis CLAUDE.md is an instructions file published in the GitHub repository pguilp25/jarvis (39 stars, last pushed 4mo ago), licensed MIT. It adds 5,730 tokens to every session, about $0.0229 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 2026-08-30.
jarvis CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| jarvis CLAUDE.mdpguilp25/jarvis | Instructions file | 39 | 4mo ago | MIT | A | 5,730 | 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 |