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/prism-shadow/agenthub/copilot-instructionsgit clone --depth 1 https://github.com/Prism-Shadow/agenthubWhat 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 | $0.01343 | $0.01343 |
| Opus 5 | $0.00672 | $0.00672 |
| Sonnet 5 | $0.00269 | $0.00269 |
| Haiku 4.5 | $0.00134 | $0.00134 |
Grade A, and why
agenthub 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.
How it starts
The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Coding Guidelines
You are a senior software engineer working on the AgentHub project.
Project Overview
AgentHub is the only SDK you need to connect to state-of-the-art LLMs.
Repository Structure
-
src_py/- Python implementationagenthub/- Main Python packagepyproject.toml- Python project configurationMakefile- Python build and test commandstests/- Python test files
-
src_ts/- TypeScript implementationsrc/- TypeScript source filespackage.json- Node.js package configurationtsconfig.json- TypeScript compiler configurationMakefile- TypeScript build and test commandstests/- TypeScript test files
-
llmsdk_docs/- Reference documentation for AI model SDKs- See this directory for detailed development guidelines and code conventions
Coding Standards
General Code Quality
- Avoid trivial comments: Do not add comments that simply restate what the code obviously does. Comments should explain why something is done, not what is being done when it's already clear from the code itself.
- ❌ Bad:
# Add temperaturebeforeconfig['temperature'] = 0.7 - ❌ Bad:
# Loop through itemsbeforefor item in items: - ✅ Good:
# Workaround: Claude requires max_tokens to be specifiedbeforeconfig['max_tokens'] = 1000 - ✅ Good: Comments explaining complex algorithms, non-obvious business logic, or workarounds for known issues
- ❌ Bad:
Python
- Follow the Google Python Style Guide
- Maintain Python 3.11+ compatibility
- Run
make lintandmake testfromsrc_py/before committing
TypeScript
- Use ESLint for code quality
- Follow TypeScript strict mode conventions
- Run
make lintandmake testfromsrc_ts/before committing
Implementation Rules
When adding support for new AI models in auto_client.py, follow these rules:
- DO NOT use generic matching like
if "claude" in model.lower()as this is too broad, always match models by explicit version number (e.g., claude4_5). - Put the implementation of the new model in a separate folder with the model identifier as the folder name, such as
claude4_5/for Claude 4.5 series models. - DO NOT create new files or directories in examples and tests when adding a new model, use test function parameters or environment variables instead.
- Always consult the llmsdk_docs/README.md for AI model SDK usage details.
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.
- 2d ago First seen · 110 lines · 1,343 tokens per session scan A 9cdd766587ba
agenthub copilot-instructions.md is an instructions file published in the GitHub repository Prism-Shadow/agenthub (111 stars, last pushed 6d ago), licensed Apache-2.0. It adds 1,343 tokens to every session, about $0.0067 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.
Other instructions, from other repositories
security GEMINI.md
Instructions for gemini-cli-extensions/security, covering standard operating procedures: security analysis guidelines, persona and guiding principles, skillset: permitted tools & investigation, skillset: sast vulnerability analysis and 1.1. hardcoded secrets.
jules GEMINI.md
Instructions for gemini-cli-extensions/jules, covering jules extension, command: /jules, activation and response, starting a jules task and jules-related queries.
self-command GEMINI.md
Instructions for stevenAthompson/self-command, covering self command extension, tools, 1. selfcommand, 2. geminisleep and 3. watchlog.
run-long-command GEMINI.md
Instructions for stevenAthompson/run-long-command, covering run long command extension, usage, tool signature, example and how it works.
GeminiSharpSDK AGENTS.md
Instructions for managedcode/GeminiSharpSDK, covering agents.md, conversations (self-learning), rules to follow (mandatory, no exceptions), commands and task delivery (all tasks).
gemini-for-claude-code CLAUDE.md
Claude Code instructions for coffeegrind123/gemini-for-claude-code, covering claude code: best practices for effective collaboration, task management, file handling and reading, file editing and reliable code insertion with multiedit.