orcbot copilot-instructions.md

Repository instructions for OrcBot, a TypeScript and Node.js AI agent that chats through services such as Telegram, WhatsApp, Discord, and the web.

In plain words
What is it for?
Use them when modifying OrcBot’s command line, decision engine, prompts, simulation, parser, context handling, channels, or memory.
Why use it?
They map the agent’s architecture and file-backed memory so code changes fit its planning, tool execution, parsing, and storage systems.

Instructions file for GitHub Copilot

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/fredabila/orcbot/copilot-instructions
Clone the repo
git clone --depth 1 https://github.com/fredabila/orcbot

Made for: GitHub Copilot.

Per session 4,142 This file is loaded in full into every session.
When invoked 4,142 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 $0.04142 $0.04142
Opus 5 $0.02071 $0.02071
Sonnet 5 $0.00828 $0.00828
Haiku 4.5 $0.00414 $0.00414

Measured 2d ago against content hash 451c0dc71df3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

orcbot copilot-instructions.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.

.github/copilot-instructions.md · 314 lines

How it starts

The opening of the file, as written. The whole thing — 314 lines — stays where its author put it; the contents beside it link to each section on GitHub.

OrcBot AI Coding Instructions

Project Overview

OrcBot is a TypeScript/Node.js autonomous AI agent that can chat over Telegram, WhatsApp, Discord, and a web gateway. It plans multi-step tasks, executes tools (skills), browses the web, and learns from its interactions. All state is file-backed under ~/.orcbot/ by default.

Architecture Map

Layer File(s) Role
CLI entrypoint src/cli/index.ts Wires all subsystems, TUI menus, user commands
Core agent src/core/Agent.ts Orchestrates memory, LLM, skills, channels, scheduling; runs the main action loop
Decision engine src/core/DecisionEngine.ts Assembles context + prompt, calls LLM, validates response, runs termination review
Modular prompts src/core/prompts/ 8 task-specific helpers + PromptRouter; only relevant helpers are injected per task
Simulation/planning src/core/SimulationEngine.ts Produces a pre-plan before execution starts
Parser src/core/ParserLayer.ts Normalizes raw LLM text into structured JSON (3-tier fallback)
Context compactor src/core/ContextCompactor.ts Truncation and LLM-based summarization for oversized context
Decision pipeline src/core/DecisionPipeline.ts Post-parse guardrails on tool calls
Memory src/memory/MemoryManager.ts Short/episodic/long memory in JSON, consolidation, daily markdown logs
Action queue src/memory/ActionQueue.ts Priority-sorted durable queue with retry, chaining, TTL, stale recovery
Storage src/storage/JSONAdapter.ts Atomic JSON persistence with backup/recovery
LLM routing src/core/MultiLLM.ts Routes by model name (gemini→Google, nvidia:→NVIDIA, else OpenAI), auto-fallback
Skills manager src/core/SkillsManager.ts Registry, dynamic plugin loading, skill matching
Channels src/channels/ Telegram (Telegraf), WhatsApp (Baileys), Discord (discord.js), Gateway (Express+WS)
Browser src/tools/WebBrowser.ts Playwright with semantic snapshots, Serper search, 2Captcha
Scheduler src/core/Scheduler.ts Cron-based via croner, emits scheduler:tick on EventBus
Config src/config/ConfigManager.ts YAML hot-reload, feature toggles, API keys
Token tracking src/core/TokenTracker.ts Per-model token/cost tracking and budgets
Runtime tuner src/core/RuntimeTuner.ts Auto-adjusts limits based on runtime signals

Read the full file on GitHub · 314 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 · 314 lines · 4,142 tokens per session scan A 451c0dc71df3

Subscribe to this mod's changes

orcbot copilot-instructions.md is an instructions file published in the GitHub repository fredabila/orcbot (11 stars, last pushed 5mo ago), licensed MIT. It adds 4,142 tokens to every session, about $0.0207 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.