export_context_bundle
Export a portable Markdown or JSON bundle of structured memory for cross-tool context transfer, audits, or resuming work. Supports prime, query, and graph selection modes with filtering by agent, project, or session.
Instructions
Export a portable Markdown and/or JSON context bundle for handing memory to another AI or a human. Use for cross-tool context transfer, audits, and resumable work. Returns file paths, counts, and render hints.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| mode | No | Bundle selection mode: scoped prime context, query result, or broad graph export. | prime |
| query | No | Natural-language query used when mode is 'query'. | |
| agent_id | No | Optional agent or client identifier used to partition memory. | |
| project | No | Optional project or workspace name used to partition memory. | |
| session_id | No | Optional conversation or run identifier used to partition memory. | |
| max_nodes | No | Maximum number of nodes to include. | |
| max_depth | No | Relationship traversal depth for query or prime context modes. | |
| retrieval_mode | No | Retrieval strategy for query mode: graph-only, transcript replay, or fused results. | graph |
| format | No | Output format to write. | both |
| output_path | No | Optional destination file path or directory prefix for the bundle. | |
| include_edges | No | Whether relationship edges should be included in the export. | |
| include_timestamps | No | Whether created and updated timestamps should be included. | |
| include_source_prompt | No | Whether original source prompts should be included when available. | |
| audience | No | Target audience used to tune bundle rendering. | llm |