| name | kg-context |
| skill_type | skill |
| description | Stores and retrieves session-scoped context blobs in the KG so work survives across turns/agents. Use to persist or recall working context — "remember this for the session", "store/get context", "list session context". |
| license | MIT |
| tags | ["graph-os","context","session"] |
| tier | core |
| metadata | {"author":"Genius","version":"0.1.0"} |
kg-context
Condensed intent-surface note (Seam 8). Under the default intent surface (MCP_TOOL_MODE=intent), graph_context is held back from the default tool list (nothing removed — REST + _execute_tool still reach it exactly as documented below). Two ways to use this skill unchanged: (1) load_tools(tools=["graph_context"]) once per session (as below), then proceed exactly as documented; or (2) call the ask intent verb with the same natural-language request — the resolver routes to graph_context for you and returns the result plus a routing justification. Set MCP_TOOL_MODE=condensed/verbose/both to expose the granular tools eagerly instead.
graph_context is a small session-anchored key/value store backed by ContextBlob nodes. Actions: put (store content under a session_id/key, optional ttl_s), get (by context_id), list (by session). Stored blobs are linked to a Session node for reliable id-anchored retrieval.
Invoke
- MCP:
load_tools(tools=["graph_context"]), then graph_context(action="put", content="...", session_id="s1").
- REST twin:
POST /graph/context with {"action": "put", "content": "...", "session_id": "s1"}.
Example
graph_context(action="put", content="user prefers efficient mode", session_id="s1", key="prefs")
graph_context(action="list", session_id="s1")
