AITM keeps a fair amount of state locally so it can operate reliably and remember what it has learned about your projects. This is a plain-language map of what's stored, and why it matters to you.

What Data AITM Stores

The data map

What it is	Where it lives	Why it matters
Per-project database	A dedicated database file for each managed project	Holds that project's contracts and task history — kept separate per project, so one project's data never mixes with another's
Global database	One shared database for AITM itself	Holds application-wide settings, hints, the variable registry, and other cross-project configuration
Task execution logs	Per-project log folders	The detailed, line-by-line record of what happened while a task ran — useful for debugging a task that behaved unexpectedly
Task data folders	Per-project task folders, one subfolder per task number	Contains that task's result files, metadata, and any screenshots captured for it — this powers the Results & Archive view
Worktrees	Temporary Git checkouts, one per running task	The isolated working copy a task's AI agents actually edit; created when a task starts and removed automatically once it's merged — see The Pipeline
Knowledge graph documents	Auto-generated documentation files	Human-readable exports of the contracts and dependency knowledge described in Knowledge Sharing; regenerated automatically after every merge

Why local, persisted state matters

None of this is incidental bookkeeping — it's the mechanism behind everything covered elsewhere in this guide:

• The per-project database is what lets contracts, task history, and precedent lookups survive between sessions and between tasks, instead of resetting every time you close the app.
• Task data folders are what make the Archive a genuinely useful audit trail rather than a list of titles — every result, screenshot, and metadata field you see there is read straight from these folders.
• Worktrees are the mechanical foundation of the isolation guarantee described in The Pipeline — because each task gets a truly separate filesystem checkout, in-progress work can never leak into your main branch by accident.
• The knowledge graph documents are what turn "the AI happened to remember something" into "the project has a durable, inspectable record of its own behavior" — see Knowledge Sharing.