Evaluate Cognee itself from local telemetry: recall hit rate, latency, saves, injected context, tokens saved, a scorecard and grade, before vs after an update, and why a memory was or wasn't injected last turn.
The English summary is being prepared. Showing the original description.
See how to installReview the instructions the agent will receive before you install.
Use the bundled Python 3.9+ evaluator, ported from team-skills PR 3. It reads local plugin logs and Claude transcripts and produces aggregates. It does not send data anywhere unless an explicit export or server option is requested.
For “how did Cognee perform?”, report the current session and the last seven days. Default to seven days when no range is supplied; use the user's range when given.
python3 "${CLAUDE_PLUGIN_ROOT}/skills/cognee-report/scripts/plugin_eval.py" report --since 7d --compare
Use --session <cognee-session-id> --since all for the current session. Resolve the
session from the current hook context or the host's entry in
~/.cognee-plugin/claude-code/sessions/<host-session-id>.json; do not assume a machine-wide last-recall
marker belongs to this terminal. When present, read this host session's
sessions/<host-session-id>/stats.json for exact injected/deduped counters.
--format json exposes the complete scorecard. --no-transcripts is faster but
cannot measure transcript-only latency and token usage. markers lists available
date ranges and update times; --since last-update --compare evaluates an update.
Present the computed hit rate, recall and SessionStart latency, saves, improves, injected characters, and grade. Quote evidence from the output, including sample counts. Mark absent or low-sample metrics as unmeasured. Grades are withheld when gates fail or evidence is insufficient; never convert that into a positive grade. Token counts are estimates at four characters per token. Older audit logs are clipped; say so if exact context counters are unavailable. A recalled or injected item is not proof that it changed an answer.
For controlled comparisons of memory behavior, use the separate outcomes command
and OUTCOMES.md. It runs repeated matched task trials and reports
task success, citation correctness, false-memory adoption, tokens/cost and P50/P95
latency. Keep the 45-second cloud recall limit. Distinguish the offline smoke
adapter from live model results; missing usage is unmeasured, not zero.
Keep recommendations tied to measured failures. Compare equal windows when possible. Do not infer seven-day post-release results from a short smoke test. Do not use export flags unless the user asks to export the report.
Exact current-session and recent session totals are also available without reading transcripts:
python3 "${CLAUDE_PLUGIN_ROOT}/scripts/_session_stats.py" --session <host-session-id> --days 7
These totals cover sessions that started within the window; use the evaluator's event timestamps for exact calendar-window recall metrics.
For why a specific memory was or was not injected on the last prompt, inspect its body-free receipt:
python3 "${CLAUDE_PLUGIN_ROOT}/scripts/cognee-evidence.py" explain --session-key <host-session-id>
Report per-item reasons (injected, already_in_context, over_budget,
own_session, or source availability status), retrieval rank and priority.
Versioned citations can be expanded through the search skill. Unaddressable
legacy/guidance entries have hashes only; do not present those as source citations.
The receipt describes delivery selection, not whether an item improved the answer.
The default report leads with What Cognee did for you, Where it helped,
Quality, then the unchanged grade. Omit the latency bar and the entire
Needs attention section. Keep latency measurements and gate results in the
scorecard, including low-n markers. Show the generated bars rather than inventing scores. Use --ascii for #/- bars or --format markdown
for fenced bars. --full restores the detailed layout and all recommendation
why/do lines. Session hashes are evidence references, never memory content;
new-token counts from clipped historical bodies are estimates. Delivery quality
is not outcome quality, even when a last-turn receipt exists.
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.