agentmods.dev

code-hacker AGENTS.md

Always-on working instructions for a coding-agent repository, covering its goals, important files, architecture, and default development habits.

In plain words
What is it for?
Use them as a starting guide when working on the repository's Python configuration, server scripts, MCP services, agent definitions, or development commands.
Why use it?
They encourage small reviewable changes, repeatable workflows, and checking the source of truth before changing servers, ports, endpoints, or tools.

Instructions file for CodexOpenCode

54 1mo ago A scan 1,222 tokens original MIT
View the source file hackerlibs/code-hacker
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/hackerlibs/code-hacker/agents-md
Clone the repo
git clone --depth 1 https://github.com/hackerlibs/code-hacker

Made for: Codex, OpenCode.

Per session 1,222 This file is loaded in full into every session.
When invoked 1,222 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 $0.01222 $0.01222
Opus 5 $0.00611 $0.00611
Sonnet 5 $0.00244 $0.00244
Haiku 4.5 $0.00122 $0.00122

Measured 2d ago against content hash 03e647921130, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-hacker AGENTS.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.

AGENTS.md · 111 lines

How it starts

The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md — How to work in this repo (for coding agents + humans)

This file is the repo's always-on working agreement. If you are an agent, follow it by default unless a task explicitly overrides it.

North Star

  • Make small, reviewable changes. Prefer tiny diffs that are easy to reason about.
  • Keep workflows repeatable. If a step matters, document it or script it.
  • Don't leak secrets. Never commit credentials, tokens, API keys, or .env files.

Read Before You Act

Before changing anything, locate the source of truth:

  1. Docs / run commands
    • README.md
    • start_servers.sh (server lifecycle management)
  2. Build configuration
    • pyproject.toml (Python deps + build config)
    • subagents.yaml (DeepAgent subagent definitions)
  3. Agent definitions
    • code-hacker.agent.md (VS Code Custom Agent system prompt + tool bindings)

If a task involves server ports, MCP endpoints, or tool names, read the relevant source file first and avoid "guess changes".

Repo Map (high level)

This section is a convenience only. If it gets out of date, trust the repo tree.

  • filesystem.py — MCP Server 1: File CRUD, search, command exec (port 8001)
  • git_tools.py — MCP Server 2: Git operations (port 8002)
  • code_intel.py — MCP Server 3: AST analysis, symbols, dependency graph (port 8003)
  • memory_store.py — MCP Server 4: Persistent memory + scratchpad + QA experience (port 8004)
  • code_review.py — MCP Server 5: Code quality review + health score + ydiff structural diff (port 8005)
  • multi_project.py — MCP Server 6: Multi-project workspace (port 8007)
  • lib/ — Internal engines (ydiff AST diff)
  • web_app.py — DeepAgent web interface (FastAPI + WebSocket)
  • tui_app.py — Claude Code-style TUI (Rich + prompt_toolkit)
  • subagents.yaml — 4 subagent definitions
  • static/ — Hacker-style web terminal UI
  • tests/ — LLM-powered integration test scenarios
  • start_servers.sh — Start/stop/status/restart all 6 MCP servers
  • codehacker/ — Package directory

Read the full file on GitHub · 111 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 · 111 lines · 1,222 tokens per session scan A 03e647921130

Subscribe to this mod's changes

code-hacker AGENTS.md is an instructions file published in the GitHub repository hackerlibs/code-hacker (54 stars, last pushed 1mo ago), licensed MIT. It adds 1,222 tokens to every session, about $0.0061 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 2026-08-30.

Data and AILLMLanguagesPlanningaicodergptprompt-engineeringvscode-extension
Related

Other instructions, from other repositories

helicone CLAUDE.md

Instructions for Helicone/helicone, covering claude.md, project overview, core architecture, services and data flow.

Helicone/helicone · 4,478 tokens

helicone AGENTS.md

Instructions for Helicone/helicone, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.

Helicone/helicone · 791 tokens

vscode-prompt-studio AGENTS.md

AGENTS.md instructions for Ge0rg3/vscode-prompt-studio, covering prompt studio - agents guide, general info, docs are a knowledge base, what earns a line and style.

Ge0rg3/vscode-prompt-studio · 6,773 tokens

prompt-tower CLAUDE.md

Claude Code instructions for backnotprop/prompt-tower, covering development guide, architecture overview, source directory structure, /api and /models.

backnotprop/prompt-tower · 429 tokens

apm python.instructions.md

Python development guidelines.

microsoft/apm · 78 tokens

AstrBot copilot-instructions.md

Copilot instructions for AstrBotDevs/AstrBot, covering astrbot development instructions, working effectively, bootstrap and install dependencies, running the application and dashboard build (vue.js/node.js).

AstrBotDevs/AstrBot · 736 tokens