Agentic Engineering Framework is an open-source orchestration harness and governance layer developed by Dimitri Geelen. It is designed to constrain, supervise, and track terminal-based AI coding agents (such as Claude) across multi-step software engineering tasks. The framework operates via shell-level hooks, repository governance scripts, and PreToolUse enforcement gates. In day-to-day development, it enforces task-tracked workflows where every agent session is bound to structured task IDs, session-scoped focus files, and observation logs. It validates agent actions in real time, preventing uncommitted regressions or out-of-scope file modifications before commands or git operations execute. What distinguishes this harness is its focus on deterministic engineering governance rather than unconstrained autonomy. It introduces guardrails against concurrent worktree overwrites, vendor synchronization checks to prevent governance regressions, and explicit state verification for CLI agent workflows. The project is available as an open-source repository on GitHub, operating directly within local development environments and git repositories without requiring external SaaS orchestration.
PreToolUse Governance Gates — Intercepts and validates agent tool calls via executable scripts to block unauthorized or out-of-scope commands. - Structured Task Tracking — Enforces session focus using formal task identifiers and tracked focus files. - Linked Worktree Isolation — Blocks accidental cross-boundary writes and governance file corruption across concurrent git worktrees. - Vendor & Policy Self-Checking — Runs automated parity and anti-regression checks to ensure framework governance rules remain intact during agent execution. - Episodic Observation Logging — Records structured observation entries and gap reports directly alongside session state for post-run auditability.
Use cases
Use Case: Guardrailed Agent Execution Developers running terminal-based coding agents can enforce strict pre-execution checks, ensuring agents operate only on designated focus files and satisfy verification conditions before modifying codebase state.
Use Case: Multi-Task Git Worktree Governance Engineers managing parallel tasks across linked git worktrees can prevent agents from clobbering root configurations, uncommitted tracking files, or baseline policies.
Use Case: Task State and Audit Tracking Teams executing multi-step automated refactors can maintain persistent observation logs, gap tracking, and formal verification gates throughout the task lifecycle.
Who it is for
This framework is intended for software engineers and maintainers seeking rigorous, deterministic guardrails and script-level governance over CLI coding agents. It is not intended for non-technical users or developers looking for a plug-and-play zero-configuration chat UI.
Install guide
Installation Clone the repository into your workspace or integrate it as a submodule within your project: bash git clone https://github.com/DimitriGeelen/agentic-engineering-framework.git cd agentic-engineering-framework
First Run Ensure governance gate scripts have execute permissions: bash chmod +x .agentic-framework/*.sh
Example Session Run the active task check to verify session governance and focus before launching an agent workflow: bash ./.agentic-framework/check-active-task.sh
SWE-agent — SWE-agent is an open-source autonomous software engineering agent harness built and maintained by researchers from Princeton University and Stanford University.…
AgentLint — AgentLint is an open-source repository linting, validation, and workflow enforcement harness developed by 0xmariowu. It is designed to enforce structural…
Weaver — Weaver is an open-source agent harness and orchestration framework maintained by sean35mm on GitHub. Built on top of the Bun…
Context Bridge — Context Bridge is an open-source CLI agent harness and context orchestration utility developed by serdardb. It is built to facilitate…
Hivelore — Hivelore is an open-source agent harness and contextual memory orchestrator created by Doucs91. It provides persistent project memory, decision tracking,…
Agent Runbook — Agent Runbook is an open-source contract-based multi-agent skill framework developed by KnoxOps. It acts as an orchestrator and compiler that…