Governance framework for AI coding agents — enforces task traceability, structural gates, session continuity, and audit trails for Claude Code, Cursor, and Copilot.

> Better context, better agents. Context Fabric captures history. Component Fabric maps blast radius. One rule: nothing runs without a task.
> The better the instruction — and the context around it — the higher the likelihood of a good result.
I built this because I recognised a pattern. In 25 years of enterprise IT governance — transition management at Shell, operational readiness for infrastructure programmes — the same structural requirements appear every time a powerful actor operates in a shared environment: traceability, approval gates, session continuity, failure memory. The domain changed from human operators to AI agents. The principle did not.
## What This Has Actually Stopped
**Agent tries to edit a file without a task:**
```
BLOCKED: No active task. Framework rule: nothing gets done without a task.
To unblock:
1. Create a task: fw task create --name '...' --type build --start
2. Set focus: fw context focus T-XXX
Attempting to modify: src/api/routes.ts
Policy: P-002 (Structural Enforcement Over Agent Discipline)
```
**Agent tries to force-push:**
```
══════════════════════════════════════════════════════════
TIER 0 BLOCK — Destructive Command Detected
══════════════════════════════════════════════════════════
Risk: FORCE PUSH overwrites remote history — may destroy teammates' work
Command: git push --force origin main
To proceed (after the human approves):
./bin/fw tier0 approve
Then retry the same command.
══════════════════════════════════════════════════════════
```
**Agent tries to disable its own guardrails:**
```
BLOCKED: Cannot modify .cursor/settings.json — this controls enforcement hooks.
Modifying this file could disable task gates, Tier 0 checks, and budget enforcement.
Changes to hook configuration require human review.
```
**Context running out mid-session:**
```
══════════════════════════════════════════════════════════
SESSION WRAPPING UP (~170000 tokens)
════Agent that generates comprehensive documentation, API references, architecture diagrams, and developer onboarding guides from existing code.
Agent configuration for systematic bug investigation that traces issues from error logs through the codebase to root cause with suggested fixes.
Agent for integrating third-party APIs including SDK setup, type generation, error handling, retry logic, and rate limit management.
Cursor's built-in autonomous coding agent that can make multi-file edits, run terminal commands, search the codebase, and iteratively build features with minimal human intervention.
Cloud-based autonomous coding agent that runs in the background on remote sandboxed environments, handling complex multi-step tasks while you continue working.
Cursor's multi-file editing agent within Composer mode that can create, edit, and delete files across your entire project in a single conversation.