Borrowing it
Nothing to install: this file belongs to Navarre-AI/pythia. 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/Navarre-AI/pythia/main/AGENTS.mdgit clone --depth 1 https://github.com/Navarre-AI/pythiaWrote 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/navarre-ai/pythia/agents-md)<a href="https://agentmods.dev/instructions/navarre-ai/pythia/agents-md"><img src="https://agentmods.dev/badge/instructions/navarre-ai/pythia/agents-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/navarre-ai/pythia/agents-md"><img src="https://agentmods.dev/badge/instructions/navarre-ai/pythia/agents-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 | $0.02732 | $0.02732 |
| Opus 5.5 | $0.01093 | $0.01093 |
| Sonnet 5.5 | $0.00546 | $0.00546 |
| Haiku 4.5 | $0.00273 | $0.00273 |
Grade C, and why
pythia AGENTS.md scanned grade C with 2 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 24d 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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -fsSL https://navarre.ai/get/pythia | sh -s -- --name "<their org or database>" --yes Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -fsSL https://navarre.ai/get/pythia | sh -s -- --name "<their org or database>" --yes How it starts
The opening of the file, as written. The whole thing — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Installing Pythia (for an AI coding agent)
You are an AI agent (e.g. Claude Code) installing Pythia for a user who may not be technical. They will say something like "install Pythia and connect it to my FileMaker database." Your job is to carry them from an empty machine to a working Pythia URL with as little burden on them as possible. Do the terminal work yourself; ask the user only for the few things you genuinely cannot get on your own.
Pythia is a conversational reporting app for FileMaker data: point it at a FileMaker Server, pick tables, ask questions in your own words, get real numbers, charts, and reports. No plugins, no FileMaker layouts to build. The installer sets up the app on Fly.io; everything else (the FileMaker connection, the AI key, the tables) is configured in the browser afterward.
How to run this install (behavioral rules)
- Fewest possible commands. Every command you run may cost the user an
approval click. The ENTIRE install is one script run as one command (Step 2).
Do not re-implement its steps individually; do not "check" things the script
already handles (it installs the Fly CLI itself). A good install is 2-4
commands total:
fly auth whoami, maybefly auth login, the installer, done. Fifty approvals means you are doing it wrong. - Be autonomous. Only stop for things that are truly theirs (creating accounts, entering a credit card, browser logins, their FileMaker details).
- Gather once, don't nag. Collect everything in the checklist below in ONE pass, then proceed. Don't ask a question you can answer by running a command.
- Never guess or probe the user's FileMaker server. Do not enumerate database names or try credentials you weren't given. Use exactly the host, database, account, and password the user provides.
- Never invent the app name or reuse the author's. The Fly app is the user's; ask what to call it. Never default to "pythia," "navarre," or anything derived from this repo.
- Never commit or print secrets.
.env,data/, andfly.tomlare gitignored; keep them that way. Don't echo API keys or passwords back in plain text, and never ask the user to paste a key into the chat. - On failure, diagnose and recover using the Troubleshooting section; don't loop the same failing command. If you're truly stuck, offer to open a GitHub issue with a REDACTED diagnostic (what step failed, the error, OS + tool versions, never a key, password, or the user's server host).
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.
- 24d ago Changed · +5 lines · +275 tokens per session e62268f75109
- 1mo ago Changed · +24 lines · +350 tokens per session scan A → C 2cbf5c14c0b9
- 2mo ago First seen · 102 lines · 2,107 tokens per session scan A e4246a744c68
pythia AGENTS.md is an instructions file published in the GitHub repository Navarre-AI/pythia (15 stars, last pushed 10d ago), licensed MIT. It adds 2,732 tokens to every session, about $0.0109 per session on Opus 5.5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.