Fli is a Python library, command-line tool, and MCP server for searching Google Flights data programmatically. Travelers and software agents use it to search flights, compare dates, filter results, and find cheaper travel options. The catalogue entries provide instructions and a skill for using Fli.
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/punitarani/fli/claude-mdgit clone --depth 1 https://github.com/punitarani/fliWrote 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/punitarani/fli/claude-md)<a href="https://agentmods.dev/instructions/punitarani/fli/claude-md"><img src="https://agentmods.dev/badge/instructions/punitarani/fli/claude-md.svg" alt="Measured on agentmods" 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 | $0.02716 | $0.02716 |
| Opus 5 | $0.01358 | $0.01358 |
| Sonnet 5 | $0.00543 | $0.00543 |
| Haiku 4.5 | $0.00272 | $0.00272 |
Grade A, and why
fli CLAUDE.md scanned grade A with 1 finding 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 5d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- Rate-limited HTTP client (10 req/sec) using curl-cffi for browser impersonation How it starts
The opening of the file, as written. The whole thing — 219 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.
Project Overview
Fli is a Python library that provides programmatic access to Google Flights data through direct API interaction (reverse engineering). The project consists of:
- CLI interface (
fli/cli/) - Typer-based command line tool withflightsanddatescommands - MCP server (
fli/mcp/) - Model Context Protocol server for AI assistant integration - Core utilities (
fli/core/) - Shared parsing and building utilities - Search engine (
fli/search/) - Flight and date search implementations using Google Flights API - Data models (
fli/models/) - Pydantic models for airports, airlines, and flight data structures
Development Commands
Core Development Tasks
# Install dependencies
uv sync --all-extras
# Run tests (use these specific commands)
make test # Standard test suite
make test-fuzz # Run fuzzing tests (pytest -vv --fuzz)
make test-all # Run all tests (pytest -vv --all)
uv run pytest -vv # Alternative direct command
# Code quality
make lint # Check code with ruff
make lint-fix # Auto-fix linting issues
make format # Format code with ruff
uv run ruff check . # Direct ruff check
uv run ruff format . # Direct ruff format
# MCP server
fli-mcp # Run MCP server on STDIO
fli-mcp-http # Run MCP server over HTTP
# Documentation
make docs # Build MkDocs documentation
uv run mkdocs serve # Serve docs locally
uv run mkdocs build # Build static docs
Test Configuration
- Tests use pytest with custom markers:
fuzz(requires--fuzzflag) andparallel(for pytest-xdist) - Test structure mirrors source code:
tests/cli/,tests/models/,tests/search/,tests/mcp/ - Fuzzing tests are available but gated behind
--fuzzflag
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.
- 5d ago First seen · 219 lines · 2,716 tokens per session scan A 37c4a240bc77
fli CLAUDE.md is an instructions file published in the GitHub repository punitarani/fli (3,135 stars, last pushed 1mo ago), licensed MIT. It adds 2,716 tokens to every session, about $0.0136 per session on Opus 5. A static security scan graded it A with 1 finding (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
anytype-mcp CLAUDE.md
Instructions for anyproto/anytype-mcp, covering claude.md, common commands, development, testing and code quality.
apitap CLAUDE.md
Instructions for n1byn1kt/apitap, covering claude.md, using apitap as an mcp tool, commands, architecture and module map.
apitap copilot-instructions.md
Instructions for n1byn1kt/apitap, covering copilot instructions for apitap, build and test commands, high-level architecture and key repository conventions.
StitchAPI CLAUDE.md
Claude Code instructions for rejifald/StitchAPI, covering stitchapi — working agreement for claude, edit in a fresh worktree off main, nodemodules in a fresh worktree and using the stitchapi docs mcp.
apifable AGENTS.md
AGENTS.md instructions for ycs77/apifable, covering agents.md, keep this file strict, constraints, commands and rules.
FFBB-MCP-Server AGENTS.md
Instructions for nickdesi/FFBB-MCP-Server, covering ffbb mcp server, langue, persona, workflow ffbb (outils mcp) and ressources mcp.