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/dimitrigeelen/agentic-engineering-framework/agents-mdgit clone --depth 1 https://github.com/DimitriGeelen/agentic-engineering-frameworkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/dimitrigeelen/agentic-engineering-framework/agents-md)<a href="https://agentmods.dev/instructions/dimitrigeelen/agentic-engineering-framework/agents-md"><img src="https://agentmods.dev/badge/instructions/dimitrigeelen/agentic-engineering-framework/agents-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.01270 | $0.01270 |
| Opus 5 | $0.00635 | $0.00635 |
| Sonnet 5 | $0.00254 | $0.00254 |
| Haiku 4.5 | $0.00127 | $0.00127 |
Grade C, and why
agentic-engineering-framework AGENTS.md scanned grade C with 1 finding 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 today.
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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
4. **Destructive commands require approval.** Force push, hard reset, `rm -rf`, DROP TABLE → `fw tier0 approve`. How it starts
The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — Agentic Engineering Framework
Standard agent instruction file (AAIF/Linux Foundation convention). For Claude Code-specific integration, see
CLAUDE.md. For the full provider-neutral guide, seeFRAMEWORK.md.
Identity
- Name: Agentic Engineering Framework
- Purpose: Governance framework for AI coding agents — enforces task traceability, structural gates, session continuity, and audit trails.
- Version: 1.0.0
Core Principle
Nothing gets done without a task. This is enforced structurally, not by agent discipline.
Rules
- Create a task before editing files. Use
fw work-on "description" --type buildto create and focus. - Every commit references a task. Format:
T-XXX: description. Enforced by commit-msg hook. - Never bypass structural gates. If a gate blocks you, stop and ask the human.
- Destructive commands require approval. Force push, hard reset,
rm -rf, DROP TABLE →fw tier0 approve. - Generate a handover before ending. Use
fw handover --committo preserve session context.
Authority Model
Human → SOVEREIGNTY → Can override anything
Framework → AUTHORITY → Enforces rules, checks gates
Agent → INITIATIVE → Can propose, never decides
Quick Commands
| Action | Command |
|---|---|
| Start work | fw work-on "name" --type build |
| Commit | fw git commit -m "T-XXX: description" |
| Run audit | fw audit |
| Health check | fw doctor |
| End session | fw handover --commit |
| Check status | fw context status |
Enforcement Tiers
| Tier | Scope | Bypass |
|---|---|---|
| 0 | Destructive commands | Human approval via fw tier0 approve |
| 1 | All file modifications | Create a task first |
| 2 | Situational exceptions | Single-use, mandatory logging |
| 3 | Read-only operations | Pre-approved |
Memory System
The framework maintains three memory layers:
- Working memory (
.context/working/) — Current session state - Project memory (
.context/project/) — Patterns, decisions, learnings - Episodic memory (
.context/episodic/) — Completed task histories
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.
- today First seen · 114 lines · 1,270 tokens per session scan C 398f9d37e810
agentic-engineering-framework AGENTS.md is an instructions file published in the GitHub repository DimitriGeelen/agentic-engineering-framework (13 stars, last pushed today), licensed Apache-2.0. It adds 1,270 tokens to every session, about $0.0064 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other instructions, from other repositories
synkb CLAUDE.md
Claude Code instructions for Syntax753/synkb, covering claude.md, read these first, rules that are easy to get wrong, shell and testing.
martin-loop AGENTS.md
AGENTS.md instructions for Keesan12/martin-loop, covering martinloop agent operating rules, repository authority, mandatory internal-first sequence, prohibited actions and required internal health proof.
orloj AGENTS.md
Instructions for OrlojHQ/orloj, covering orloj agent instructions, must-follow sync rules and working style.
Alex_Skill_Mall mall-maintenance-rules.instructions.md
Always-on routing for Mall maintenance work — fires the right Mall skill at the right moment. Distinguishes Mall-owned automation from out-of-scope editorial work.
Bratan CLAUDE.md
Claude Code instructions for AllanWessels/Bratan, covering rag refiner, what this project is, repo layout, how to work in this repo and run the loop.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).