Convert gitgram/ and aitimemachine/ from embedded git repos (which showed as opaque submodule gitlinks on GitHub) into fully-tracked first-party content. AITM now merges its 7 vendored repos as first-party engines: - memory: beads (Go task DAG) + context7 (docs SDK), unified into saved_contexts - git: vendored C git (bundled binary) + gitnexus (in-process code intelligence) - agentic: a2a (@a2a-js/sdk) + pal (AITM-managed Python subprocess) Includes the full session's product build: Gitgram web/mobile/backend features, AITM website/CLI/MCP, the unified memory layer, engine build tooling (scripts/build-engines.sh, Dockerfile, CI), docs (README, ENGINES.md, ARCHITECTURE.md), and extensive test suites. Build artifacts, node_modules, external/, and gemini-cli are gitignored. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
5.8 KiB
AITimeMachine API Reference
TimeMachine Class
Constructor
new TimeMachine(projectPath: string)
Methods
initialize(): Promise<void>
Initializes the time machine, setting up git repo and loading existing data.
createSnapshot(filePath?: string): Promise<void>
Creates a timestamped snapshot of the current code state and conversation.
revertToSnapshot(index: number): Promise<void>
Reverts to the snapshot at the given index, with user confirmation if configured.
getSnapshots(): Snapshot[]
Returns array of all snapshots.
getSnapshot(index: number): Snapshot | undefined
Returns a specific snapshot by index.
logConversation(role: 'user' | 'assistant', message: string): void
Logs a conversation entry.
getCurrentConversation(): ConversationEntry[]
Returns the current conversation log.
getDiff(fromIndex: number, toIndex?: number): string
Returns a diff between the snapshot at fromIndex and toIndex (or current state if not specified).
clearConversation(): void
Clears the conversation log.
startWatching(): void
Starts file watching for automatic snapshots.
stopWatching(): void
Stops file watching.
MCP Tools
Core Snapshot Tools
create_snapshot
Create a timestamped snapshot of the current code state and conversation.
- Input:
{ filePath?: string } - Output:
"Snapshot created"
list_snapshots
List all available snapshots.
- Input:
{} - Output: JSON array of snapshots with index, timestamp, conversationCount
revert_to_snapshot
Revert to a specific snapshot by index.
- Input:
{ index: number } - Output: Confirmation message
get_diff
Get diff between snapshots or current state.
- Input:
{ fromIndex: number, toIndex?: number } - Output: Text diff
Conversation Tools
log_conversation
Log a conversation entry to persist AI memory.
- Input:
{ role: 'user' | 'assistant', message: string } - Output: Confirmation message
get_conversation
Get the current conversation log.
- Input:
{} - Output: JSON array of conversation entries
search_history
Search conversations and code across all snapshots.
- Input:
{ query: string, type?: 'all' | 'conversation' | 'code' } - Output: JSON array of matching snapshots with previews
Cloud Sync Tools
cloud_sync
Sync snapshots with cloud storage (push and pull).
- Input:
{} - Output: Sync results (pushed/pulled counts)
cloud_push
Push local snapshots to cloud.
- Input:
{} - Output: Upload count
cloud_pull
Pull snapshots from cloud.
- Input:
{} - Output: Download count
cloud_status
Get cloud sync status and configuration.
- Input:
{} - Output: JSON with configured, endpoint, projectId, lastSync, syncedCount
cloud_configure
Configure cloud sync endpoint and API key.
- Input:
{ endpoint: string, apiKey: string } - Output: Confirmation with project ID
Timeline Tools
get_timeline
Get a formatted timeline of all snapshots.
- Input:
{} - Output: Formatted text timeline
restore_by_description
Find and restore a snapshot by searching conversation history.
- Input:
{ description: string } - Output: Confirmation of restored snapshot
CLI Commands
Core Commands
| Command | Description |
|---|---|
aitm init <path> |
Initialize project |
aitm snapshot [file] |
Create snapshot |
aitm list |
List snapshots |
aitm revert <index> |
Revert to snapshot |
aitm diff <from> [to] |
Show diff |
aitm log <role> <msg> |
Log conversation |
Timeline & Search
| Command | Description |
|---|---|
aitm timeline |
Interactive TUI browser |
aitm timeline -s |
Simple text timeline |
aitm search <query> |
Search all history |
aitm search -c <query> |
Search conversations only |
aitm search -f <query> |
Search code only |
aitm restore <desc> |
Find & restore by description |
Cloud Sync
| Command | Description |
|---|---|
aitm sync |
Sync with cloud |
aitm sync-config <url> <key> |
Configure cloud |
aitm sync-status |
Show sync status |
aitm export <file> |
Export to file |
aitm import <file> |
Import from file |
UI & Integration
| Command | Description |
|---|---|
aitm ui |
Start web UI |
aitm ui -p <port> |
Web UI on custom port |
aitm watch <cmd> |
Monitor AI CLI tool |
aitm mcp |
Run MCP server |
aitm config |
Show config |
aitm set-config <k> <v> |
Update config |
Data Structures
Snapshot
interface Snapshot {
timestamp: Date;
codeState: string;
conversationLog?: ConversationEntry[];
}
ConversationEntry
interface ConversationEntry {
timestamp: Date;
role: 'user' | 'assistant';
message: string;
}
SyncConfig
interface SyncConfig {
endpoint?: string;
apiKey?: string;
projectId?: string;
autoSync?: boolean;
}
SearchResult
interface SearchResult {
snapshotIndex: number;
timestamp: Date;
matches: SearchMatch[];
}
MCP Integration Example
Add to your AI tool's MCP config:
{
"mcpServers": {
"aitimemachine": {
"command": "aitm",
"args": ["mcp"],
"cwd": "/path/to/your/project"
}
}
}
Transport
AITimeMachine’s MCP server uses stdio by default:
aitm mcp
Optional HTTP/SSE transports can be layered for custom clients, but stdio is the supported v1 path.
The AI can then use tools like:
create_snapshotbefore making changeslog_conversationto persist memorysearch_historyto recall past discussionscloud_syncto backup across devicesrestore_by_descriptionto time travel