# Terragrunt 0 89

> Apply Terragrunt 0.89 HCL, CLI, orchestration, migration, and behaviour changes. Load for Terragrunt work targeting 0.89 or later when the model's knowledge may predate the 2025-10-06 release.

- Skill: `cedws/terragrunt-0-89` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add cedws/terragrunt-0-89`
- Raw SKILL.md: https://api.skillmd.com/api/skills/cedws/terragrunt-0-89/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: cedws (https://skillmd.com/u/cedws)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/cedws/terragrunt-0-89

---


# Terragrunt 0.89

This overview applies to target Terragrunt versions of at least 0.89.

## Completed experiments

Breaking behavioural change: the runner-pool concurrency model became the default, dynamically scheduling a unit as soon as its dependencies are satisfied, or its dependants for destroy operations.

The run-report experiment became generally available. Multi-unit runs emit a summary by default; `--disable-report-summary` suppresses it, while `--report` and `--report-schema-file` support detailed CSV or JSON reporting.

Automatic provider cache directory configuration became generally available for OpenTofu 1.10 or later. Terragrunt sets `TF_PLUGIN_CACHE_DIR` for OpenTofu's concurrency-safe native cache; `--no-auto-provider-cache-dir` disables it. Terraform does not provide this capability, so Terraform workflows require a different caching choice.