hello-agent-cli CLAUDE.md

A project instruction file for the ricoNext/hello-agent-cli repository. It documents the Bun commands, environment variables, and the paths used to route requests through the command-line application.

In plain words
What is it for?
Use it when changing this repository, running its development commands, configuring its OpenAI-compatible API settings, or tracing how command-line input reaches the agent.
Why use it?
It gives coding agents the repository-specific context needed to run, format, lint, and understand the project without rediscovering its setup.

Instructions file

Install

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.

agentmods
npx agentmods add instructions/riconext/hello-agent-cli/claude-md
Clone the repo
git clone --depth 1 https://github.com/ricoNext/hello-agent-cli
Per session 647 This file is loaded in full into every session.
When invoked 647 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00647 $0.00647
Opus 5 $0.00324 $0.00324
Sonnet 5 $0.00129 $0.00129
Haiku 4.5 $0.00065 $0.00065

Measured 2d ago against content hash 0aafed0759e9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

hello-agent-cli 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 2d 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.

CLAUDE.md · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 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.

Commands

bun install          # install dependencies
bun run dev          # run CLI in development
bun x ultracite fix  # format & lint (run before committing)
bun x ultracite check # check for issues without fixing

No test suite is configured yet.

Environment Variables

Variable Required Default Description
OPENAI_API_KEY Yes API key (also works with Alibaba Qwen via custom base URL)
OPENAI_BASE_URL No OpenAI default Custom API endpoint
OPENAI_MODEL No gpt-4o Model name

Architecture

This is a Bun-based AI agent CLI with three usage modes:

  1. Single querybun run src/index.ts "prompt" calls runQuery() for a one-shot response
  2. Pipe modeecho "prompt" | bun run src/index.ts -p runs runAgentPipe(), the agentic loop
  3. Interactive REPLbun run src/index.ts (no args) launches the Ink-based terminal UI

Entry & Routing (src/index.tssrc/cli.ts)

index.ts fast-paths --version to avoid loading Commander. All other invocations delegate to cli.ts which uses Commander to route to the three modes above.

Agentic Loop (src/agent/loop.ts)

runAgentPipe() is the core agent loop:

  • Calls the model with bash tool definitions (src/tools/openai-tools.ts)
  • If the model returns tool_calls, executes them via src/tools/bash.ts (Bun subprocess, stdout/stderr capped at 8000 chars)
  • Appends tool results to message history and loops until no tool calls or --max-turns is reached

Context Building (src/agent/context.ts)

buildSystemPrompt() enriches the system prompt with:

  • Git state (branch, recent commits, working-tree diff) via getSystemContext()
  • .claude/CLAUDE.md rules and current date via getUserContext()

Results are memoized for the session lifetime.

Streaming (src/agent/stream-query.ts)

streamQuery() wraps the OpenAI streaming API and accepts an onDelta callback invoked per token — used by the REPL for live output.

Read the full file on GitHub · 62 lines

Changes

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.

  1. 2d ago First seen · 62 lines · 647 tokens per session scan A 0aafed0759e9

Subscribe to this mod's changes

hello-agent-cli CLAUDE.md is an instructions file published in the GitHub repository ricoNext/hello-agent-cli (5 stars, last pushed 3mo ago), licensed MIT. It adds 647 tokens to every session, about $0.0032 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 2026-08-31.