← Docs

Contexere MCP Server

48 evaluation tools for Claude Code, Cursor, and any MCP-compatible client. Manage agents, ground truth datasets, eval runs, deployments, and drift monitoring — without leaving your editor.

Quick Startpip install contexere-ai

1. Install

$ pip install contexere-ai

2. Set your API key

$ export CONTEXERE_API_KEY="ck_your_key"

3. Add to Claude Code

$ claude mcp add contexere -- conte mcp

Or add to project config

// .claude/.mcp.json
{
"mcpServers": {
"contexere": {
"command": "conte",
"args": ["mcp"],
"env": {
"CONTEXERE_API_KEY": "${CONTEXERE_API_KEY}"
}
}
}
}

Available Tools

11 categories · 52 tools

Agent Tools

7

Inspect and manage your instrumented agents and their version history.

list_agents

List all agents in the project with their versions, environments, and status.

get_agent

Get details for a specific agent by name or ID.

list_agent_versions

List all versions for an agent.

get_agent_version

Get a specific version of an agent including prompts.

diff_agent_versions

Show the diff between two agent versions (what changed in prompts/config).

create_agent_version

Create a new agent version with updated prompts.

link_gt_to_version

Link a ground truth dataset to an agent version for evaluation.

Ground Truth Tools

12

Create, manage, and evolve ground truth datasets that encode domain expertise.

list_gt_datasets

List ground truth datasets with optional filters.

get_gt_dataset

Get details of a ground truth dataset.

create_gt_dataset

Create a new ground truth dataset.

delete_gt_dataset

Delete a ground truth dataset and all its entries.

list_gt_entries

List entries in a ground truth dataset.

add_gt_entries

Add entries to a ground truth dataset.

update_gt_entry

Update a ground truth entry.

delete_gt_entry

Delete a single ground truth entry.

diff_gt_datasets

Compare two versions of a ground truth dataset.

get_gt_evolution

Show the version history of a GT dataset including expert enrichments.

preview_corrections

Preview expert corrections from an eval run's reviews before enriching GT.

enrich_gt

Create a new GT version enriched with expert corrections from an eval run.

Eval Tools

7

Run evaluations comparing agent versions against ground truth benchmarks.

run_eval

Run an evaluation comparing an agent version against a ground truth dataset.

list_eval_runs

List recent eval runs with optional filters.

get_eval_run

Get detailed results for an eval run.

list_eval_entries

List entries from an eval run (pass/fail per GT entry).

analyze_eval_run

Trigger AI Analyst to analyze an eval run's failures and find patterns.

generate_review_schema

Generate targeted review questions from eval failure patterns.

send_to_review

Send failed eval entries to the expert review queue.

Suggestion Tools

3

AI-generated context improvements synthesized from expert feedback.

list_suggestions

List AI-generated improvement suggestions for an eval run.

update_suggestion

Accept, reject, or edit a suggestion. Accepting creates a new agent version.

synthesize_suggestions

Trigger AI Editor to synthesize expert feedback into prompt improvement suggestions.

Lifecycle Tools

3

Promote agent versions across environments with full rollback support.

promote_agent

Promote an agent version to staging or production.

rollback_agent

Rollback an agent to its previous environment.

get_promotion_history

Get promotion/rollback history for an agent.

Performance Tools

3

Track agent quality over time and detect regressions across versions.

get_agent_performance

Get eval score timeline across all versions of an agent.

get_regressions

Detect regressions across all versions — entries that were right then wrong.

analyze_regressions

AI root cause analysis on regressions — explains WHY entries broke.

Classification & Metrics Tools

3

Confusion matrices, precision/recall tracking, and classification transitions.

get_confusion_matrix

Get confusion matrix and metrics (precision, recall, F1, accuracy) for an eval run.

get_metrics_history

Get precision/recall/F1/accuracy history across eval runs for an agent.

get_classification_transitions

Get classification transitions (TP to FN, TN to FP, etc.) from the previous eval run.

Monitoring Tools

4

Configure and check production drift detection against ground truth.

check_drift

Check production drift against ground truth.

list_drift_reports

List recent drift reports for an agent.

get_monitoring_config

Get drift monitoring configuration for an agent.

update_monitoring_config

Configure drift monitoring for an agent.

Deployment Tools

5

Full and canary deployments with validation and rollback.

create_deployment

Create a new deployment (full or canary).

validate_deployment

Run validation on a deployment (compares against GT baseline).

start_canary

Start canary traffic for a deployment.

promote_deployment

Promote a canary deployment to full rollout.

rollback_deployment

Rollback a deployment.

Query Tools

3

Search traces, feedback, and aggregate agent statistics.

query_spans

Query agent execution traces with optional filters.

query_feedback

Query expert feedback/reviews with optional filters.

get_agent_stats

Get aggregate statistics for an agent (spans, reviews, versions).

Automation Tools

2

Run N automated improvement rounds: analyst → editor → new version → re-eval.

run_automated_loop

Run N automated improvement rounds (analyst → editor → new version → re-eval).

get_loop_status

Get status of an automated improvement loop.