Search Cognee memory. Session memory is automatically searched on every prompt via hooks. Use this skill explicitly for permanent knowledge graph search, filtered category search, or when you need more results than the automatic lookup provides.
The English summary is being prepared. Showing the original description.
See how to installReview the instructions the agent will receive before you install.
Search both session memory and the permanent knowledge graph, optionally filtered by data category.
Session memory is searched automatically on every user prompt via the user-prompt-submit hook. You do not need to run this skill to access current-session context.
Knowledge is organized into three categories via node_set:
| Category | Node set | Contains |
|---|---|---|
| user | user_context | User preferences, corrections, personal facts |
| project | project_docs | Repository docs, code context, architecture decisions |
| agent | agent_actions | Tool call logs, reasoning traces, generated artifacts |
Use the cognee_recall tool to search Cognee memory. It calls the running Cognee server (POST /api/v1/recall) directly in-process — no shell, no subprocess.
One broad search is usually enough — the user-prompt-submit hook already injects session/trace/graph context every turn, so avoid running many targeted searches.
cognee_recall(query="your search query", top_k=10, scope="auto")
scope="auto" — session cache + permanent graph (default)scope="graph" — permanent graph onlyscope="session" — current session onlyCategories (user_context / project_docs / agent_actions) filter by node set. Use a direct curl to pass node_name:
curl -s -X POST "${COGNEE_BASE_URL:-http://localhost:8011}/api/v1/recall" \
-H "Content-Type: application/json" \
-H "X-Api-Key: ${COGNEE_API_KEY:-}" \
-d '{"query": "...", "top_k": 5, "only_context": true, "scope": ["graph"], "node_name": ["project_docs"]}'
The server is authoritative. If a search returns empty but you expect content, confirm directly — do not conclude "not found" from an empty result without checking:
curl -s -X POST "${COGNEE_BASE_URL:-http://localhost:8011}/api/v1/recall" \
-H "Content-Type: application/json" \
-H "X-Api-Key: ${COGNEE_API_KEY:-}" \
-d '{"query": "...", "top_k": 5, "only_context": true, "scope": ["graph"]}'
If the response is an {"error": ...} object rather than a list, the server was reachable but rejected/failed the request — that's an error, not "no results".
Results include a _source field:
"session" — from the session cache (current conversation)"graph" — from the permanent knowledge graphSession entries tagged with [category:agent] are automatic tool call logs.
| Signal | Action |
|---|---|
| Need current session context | Already automatic, no action needed |
| User explicitly says "search cognee" | cognee_recall(query="...") |
| "what does the codebase do" / "what did we do last time" | cognee_recall(query="...", scope="graph") |
| Need a specific category | use the node_name curl form above |
| Auto context insufficient | cognee_recall(query="...", scope="session") |
| Result empty but you expect content | Ground-truth via the curl above before concluding "not found" |
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Use when indexing a code repository into Cognee's code graph or querying it (callers, impact analysis, paths, endpoints) from Antigravity.
The English summary is being prepared. Showing the original description.
Use when indexing a code repository into Cognee's code graph or querying it (callers, impact analysis, paths, endpoints) from Codex.
The English summary is being prepared. Showing the original description.
Explain a repository's architecture (modules, routes, storage, services), find risky hotspots, draw a Mermaid diagram of the code graph, or review how far a change reaches, from Cognee's code graph.
The English summary is being prepared. Showing the original description.
Index a repository (local path or git URL) into Cognee's code graph and answer exact structural queries with no LLM calls: callers, dependencies, impact of changing a symbol, call paths, endpoints, re-index changes.
The English summary is being prepared. Showing the original description.
A multi-day digest of the user's own coding-agent sessions — sessions per day and project, most-edited files, and every lesson Cognee distilled into the knowledge graph in that window. Use only when the user explicitly asks for a weekly digest, a retro, a recap or report of their recent sessions, or asks for the `cognee-digest` skill by name. Not for questions about code or the current task.
The English summary is being prepared. Showing the original description.
A digest of a week or more of the user's own coding-agent sessions (default 7 days): sessions per day and project, most-edited files and the lessons Cognee learned, written up as shipped, decided and still open.
The English summary is being prepared. Showing the original description.