What is agentforge-graph AGENTS.md?

A project instruction file for AgentForge Graph, an application framework for building AI assistants. It defines the project structure, which files may be edited, and rules the code must follow.

Use it to guide changes to AgentForge tools, configuration, reasoning settings, and developer-owned files while preserving the project's ownership markers and architecture rules.

It gives coding assistants the project-specific context needed to make compatible changes. It also protects framework-managed files from accidental edits.

agentforge-graph AGENTS.md — Instructions file by Scaffoldic

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/scaffoldic/agentforge-graph/agents-md

Clone the repo

git clone --depth 1 https://github.com/Scaffoldic/agentforge-graph

Made for: Codex, OpenCode.

Per session 2,045 This file is loaded in full into every session.

When invoked 2,045 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.

Model	Per session	Once invoked
Fable 5	$0.02045	$0.02045
Opus 5	$0.01022	$0.01022
Sonnet 5	$0.00409	$0.00409
Haiku 4.5	$0.00204	$0.00204

Measured 2d ago against content hash 783536f94bb4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agentforge-graph 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 · 165 lines

How it starts

The opening of the file, as written. The whole thing — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AgentForge agent — AI assistant instructions

This project is built on AgentForge 0.3.1. Use these rules when suggesting changes. The file is read by Claude Code and any tool following the AGENTS.md convention.

Project shape (you must respect this)

Framework version: 0.3.1
Template: minimal
Project slug: agentforge-graph
LLM provider: anthropic

File ownership rules

Files starting with AGENTFORGE-MANAGED: are owned by the framework. Do not edit. Suggest changes to YAML config or developer-owned files instead.
Files starting with AGENTFORGE-FORKED: were customised by the developer. Edit normally; do not restore the marker.
Files with no marker are developer-owned. Edit normally.
Markdown documents may carry a  marker. Everything above belongs to the framework; everything below is the developer's custom section and survives upgrades.

Architecture invariants

Tools — use the @tool decorator on a typed function, OR subclass Tool. Type hints drive the input schema; do not hand-write JSON schemas.
Reasoning loop — do not edit. Configure via agentforge.yaml > agent.strategy.
LLM clients — do not import vendor SDKs directly. Use agent.providers["..."] or pass model="<provider>:<model_id>".
Memory — do not write SQL directly. Use agent.memory.put / .get / .query.
Costs — do not bypass BudgetPolicy. Every LLM call is checked. Bypassing it is a P3 (cost-safety) violation.
Run id — do not invent your own correlation id. Use current_run().run_id from agentforge_core.production.run_context.
Guardrails — input / output / tool-gate validators live in agentforge.guardrails. Custom validators implement the locked ABCs in agentforge_core.contracts.guardrails.

How to add common things

Open the runbook for the full step-by-step. Each runbook lives at docs/runbooks/NN-<topic>.md and is upgrade-safe (the framework maintains the managed section).

Read the full file on GitHub · 165 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.

2d ago First seen · 165 lines · 2,045 tokens per session scan A 783536f94bb4

Subscribe to this mod's changes

agentforge-graph AGENTS.md is an instructions file published in the GitHub repository Scaffoldic/agentforge-graph (4 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 2,045 tokens to every session, about $0.0102 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.

Data and AI Planning Research ai-agents code-analysis code-knowledge-graph code-search coding-agents developer-tools

agentforge-graph AGENTS.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

AgentForge agent — AI assistant instructions

Project shape (you must respect this)

File ownership rules

Architecture invariants

How to add common things

What this file has done since we first saw it

Other instructions, from other repositories

graphify AGENTS.md

CodeNib AGENTS.md

code-context AGENTS.md

codeTree CLAUDE.md

codeTree AGENTS.md

code-meridian AGENTS.md