Serveur MCP
ALPNAI — Agent Performance Tools
io.github.fredericmagnathy-ops/alpnai
Ce que fait ce MCP
Analyzes supplied agent workflow traces for cost, latency, retry behavior, quality gates, and performance reporting, with optional report purchases.
Capacités
Outils
10
outils
analyze_agent_latency
Latency Lab
Find slow agent workflows and retry overhead before scaling. Supply recorded attempts with task_id, workflow, variant, cost_usd, success and optional latency_ms. Returns P50/P95 of summed recorded attempt durations per task, retry counts, duration coverage and an optional P95 threshold check for each workflow and variant. Missing durations produce null percentiles, not zero latency. This measures supplied durations, not live end-to-end service latency. Free calculation; requires an active ALPNAI agent key; no payment or automatic deployment.
Lecture seule
Idempotent
Schéma d’entrée
{'type': 'object', 'title': 'ALPNAI recorded agent attempts', 'required': ['runs'], 'properties': {'runs': {'type': 'array', 'items': {'type': 'object', 'required': ['task_id', 'workflow', 'variant', 'cost_usd', 'success'], 'properties': {'success': {'type': 'boolean'}, 'task_id': {'type': 'string', 'pattern': '^(?!\\s)(?![\\s\\S]*\\s$)[^\\u0000-\\u001F\\u007F]+$', 'maxLength': 128, 'minLength': 1, 'description': 'Nonempty, no control characters or surrounding whitespace. The engine additionally enforces at most 128 UTF-16 code units.'}, 'variant': {'enum': ['baseline', 'candidate'], 'type': 'string'}, 'cost_usd': {'type': 'number', 'maximum': 10000, 'minimum': 0, 'description': 'USD cost of this attempt. At most six decimal places; the engine verifies whole micro-USD using floating-point tolerance. No multipleOf keyword is used, to avoid rejecting valid JSON decimals.'}, 'workflow': {'type': 'string', 'pattern': '^(?!\\s)(?![\\s\\S]*\\s$)[^\\u0000-\\u001F\\u007F]+$', 'maxLength': 80, 'minLength': 1, 'description': 'Nonempty, no control characters or surrounding whitespace. The engine additionally enforces at most 80 UTF-16 code units.'}, 'latency_ms': {'type': 'number', 'maximum': 86400000, 'minimum': 0, 'description': 'Recorded attempt duration in milliseconds; decimals are accepted. Omit if not measured.'}}, 'additionalProperties': False}, 'maxItems': 1000, 'minItems': 1}, 'config': {'type': 'object', 'required': [], 'properties': {'minSamples': {'type': 'integer', 'default': 30, 'maximum': 500, 'minimum': 2}, 'monthlyTasks': {'type': 'integer', 'maximum': 1000000, 'minimum': 1, 'description': 'Baseline logical tasks launched per month; the engine requires exactly one workflow if supplied.'}, 'minSuccessRate': {'type': 'number', 'default': 0.95, 'maximum': 1, 'minimum': 0}, 'maxP95LatencyMs': {'type': 'number', 'maximum': 86400000000, 'minimum': 0}, 'maxSuccessRateDrop': {'type': 'number', 'default': 0.02, 'maximum': 1, 'minimum': 0}}, 'additionalProperties': False}}, 'description': 'Recorded attempts, not prompts or secrets. Identical rows count as separate attempts. All three analysis tools accept this same input. The runtime validator additionally enforces six-decimal micro-USD precision, the same task_id belonging to one workflow, one workflow when monthlyTasks is given, and UTF-16 length limits. HTTP body limit: 512000 UTF-8 bytes.', 'additionalProperties': False}
Schéma de sortie
{'type': 'object', 'required': ['schema_version', 'measurement', 'percentile_method', 'groups'], 'properties': {'groups': {'type': 'array', 'items': {'type': 'object', 'required': ['workflow', 'variant', 'tasks', 'recorded_attempts', 'retry_attempts', 'duration_coverage', 'p50_ms', 'p95_ms', 'max_ms', 'threshold_ms', 'within_p95_threshold'], 'properties': {'tasks': {'type': 'integer', 'maximum': 1000, 'minimum': 1}, 'max_ms': {'anyOf': [{'type': 'number', 'maximum': 86400000000, 'minimum': 0}, {'type': 'null'}]}, 'p50_ms': {'anyOf': [{'type': 'number', 'maximum': 86400000000, 'minimum': 0}, {'type': 'null'}]}, 'p95_ms': {'anyOf': [{'type': 'number', 'maximum': 86400000000, 'minimum': 0}, {'type': 'null'}]}, 'variant': {'enum': ['baseline', 'candidate'], 'type': 'string'}, 'workflow': {'type': 'string', 'pattern': '^(?!\\s)(?![\\s\\S]*\\s$)[^\\u0000-\\u001F\\u007F]+$', 'maxLength': 80, 'minLength': 1, 'description': 'Nonempty, no control characters or surrounding whitespace. The engine additionally enforces at most 80 UTF-16 code units.'}, 'threshold_ms': {'anyOf': [{'type': 'number', 'maximum': 86400000000, 'minimum': 0}, {'type': 'null'}]}, 'retry_attempts': {'type': 'integer', 'maximum': 999, 'minimum': 0}, 'duration_coverage': {'type': 'number', 'maximum': 1, 'minimum': 0}, 'recorded_attempts': {'type': 'integer', 'maximum': 1000, 'minimum': 1}, 'within_p95_threshold': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}]}}, 'additionalProperties': True}, 'maxItems': 1000, 'minItems': 1}, 'measurement': {'type': 'string', 'const': 'summed_recorded_attempt_durations_per_task'}, 'schema_version': {'type': 'string', 'const': '1.0.0'}, 'percentile_method': {'type': 'string', 'const': 'nearest_rank'}}, 'additionalProperties': True}
audit_agent_costs
Spend Proof cost analysis
Free comparison of cost per successful task on supplied paired traces. Requires a sandbox agent key. No purchase, storage, external model call or automatic deployment. Success labels are supplied by the client.
Lecture seule
Idempotent
Schéma d’entrée
{'type': 'object', 'title': 'ALPNAI recorded agent attempts', 'required': ['runs'], 'properties': {'runs': {'type': 'array', 'items': {'type': 'object', 'required': ['task_id', 'workflow', 'variant', 'cost_usd', 'success'], 'properties': {'success': {'type': 'boolean'}, 'task_id': {'type': 'string', 'pattern': '^(?!\\s)(?![\\s\\S]*\\s$)[^\\u0000-\\u001F\\u007F]+$', 'maxLength': 128, 'minLength': 1, 'description': 'Nonempty, no control characters or surrounding whitespace. The engine additionally enforces at most 128 UTF-16 code units.'}, 'variant': {'enum': ['baseline', 'candidate'], 'type': 'string'}, 'cost_usd': {'type': 'number', 'maximum': 10000, 'minimum': 0, 'description': 'USD cost of this attempt. At most six decimal places; the engine verifies whole micro-USD using floating-point tolerance. No multipleOf keyword is used, to avoid rejecting valid JSON decimals.'}, 'workflow': {'type': 'string', 'pattern': '^(?!\\s)(?![\\s\\S]*\\s$)[^\\u0000-\\u001F\\u007F]+$', 'maxLength': 80, 'minLength': 1, 'description': 'Nonempty, no control characters or surrounding whitespace. The engine additionally enforces at most 80 UTF-16 code units.'}, 'latency_ms': {'type': 'number', 'maximum': 86400000, 'minimum': 0, 'description': 'Recorded attempt duration in milliseconds; decimals are accepted. Omit if not measured.'}}, 'additionalProperties': False}, 'maxItems': 1000, 'minItems': 1}, 'config': {'type': 'object', 'required': [], 'properties': {'minSamples': {'type': 'integer', 'default': 30, 'maximum': 500, 'minimum': 2}, 'monthlyTasks': {'type': 'integer', 'maximum': 1000000, 'minimum': 1, 'description': 'Baseline logical tasks launched per month; the engine requires exactly one workflow if supplied.'}, 'minSuccessRate': {'type': 'number', 'default': 0.95, 'maximum': 1, 'minimum': 0}, 'maxP95LatencyMs': {'type': 'number', 'maximum': 86400000000, 'minimum': 0}, 'maxSuccessRateDrop': {'type': 'number', 'default': 0.02, 'maximum': 1, 'minimum': 0}}, 'additionalProperties': False}}, 'description': 'Recorded attempts, not prompts or secrets. Identical rows count as separate attempts. All three analysis tools accept this same input. The runtime validator additionally enforces six-decimal micro-USD precision, the same task_id belonging to one workflow, one workflow when monthlyTasks is given, and UTF-16 length limits. HTTP body limit: 512000 UTF-8 bytes.', 'additionalProperties': False}
Schéma de sortie
{'type': 'object', 'required': ['schema_version', 'purpose', 'data_provenance', 'config', 'input_attempt_records', 'experimental_bias_control', 'automatic_deployment_authorized', 'methodology', 'workflows'], 'properties': {'config': {'type': 'object', 'required': ['minSamples', 'minSuccessRate', 'maxSuccessRateDrop'], 'properties': {'minSamples': {'type': 'integer', 'default': 30, 'maximum': 500, 'minimum': 2}, 'monthlyTasks': {'type': 'integer', 'maximum': 1000000, 'minimum': 1, 'description': 'Baseline logical tasks launched per month; the engine requires exactly one workflow if supplied.'}, 'minSuccessRate': {'type': 'number', 'default': 0.95, 'maximum': 1, 'minimum': 0}, 'maxP95LatencyMs': {'type': 'number', 'maximum': 86400000000, 'minimum': 0}, 'maxSuccessRateDrop': {'type': 'number', 'default': 0.02, 'maximum': 1, 'minimum': 0}}, 'additionalProperties': True}, 'purpose': {'type': 'string', 'const': 'cost_per_successful_agent_task_audit'}, 'workflows': {'type': 'array', 'items': {'type': 'object', 'required': ['workflow', 'baseline', 'candidate', 'comparison', 'gates', 'decision', 'automatic_deployment_authorized', 'realized_savings_usd', 'opportunity_monthly'], 'properties': {'gates': {'type': 'object', 'required': ['both_variants', 'minimum_distinct_tasks_per_variant', 'same_task_set', 'observed_success_rate', 'recorded_latency', 'lower_cost_per_successful_task', 'experimental_bias_control'], 'properties': {'both_variants': {'enum': ['pass', 'fail', 'unknown', 'not_requested'], 'type': 'string'}, 'same_task_set': {'enum': ['pass', 'fail', 'unknown', 'not_requested'], 'type': 'string'}, 'recorded_latency': {'enum': ['pass', 'fail', 'unknown', 'not_requested'], 'type': 'string'}, 'observed_success_rate': {'enum': ['pass', 'fail', 'unknown', 'not_requested'], 'type': 'string'}, 'experimental_bias_control': {'type': 'string', 'const': 'unknown'}, 'lower_cost_per_successful_task': {'enum': ['pass', 'fail', 'unknown', 'not_requested'], 'type': 'string'}, 'minimum_distinct_tasks_per_variant': {'enum': ['pass', 'fail', 'unknown', 'not_requested'], 'type': 'string'}}, 'additionalProperties': True}, 'baseline': {'anyOf': [{'type': 'object', 'required': ['tasks', 'attempts', 'retry_attempts', 'successful_tasks', 'total_cost_usd', 'cost_per_task_usd', 'cost_per_successful_task_usd', 'success_rate', 'success_rate_interval_95_wilson', 'p95_recorded_attempt_latency_per_task_ms', 'latency_complete'], 'properties': {'tasks': {'type': 'integer', 'maximum': 1000, 'minimum': 1}, 'attempts': {'type': 'integer', 'maximum': 1000, 'minimum': 1}, 'success_rate': {'type': 'number', 'maximum': 1, 'minimum': 0}, 'retry_attempts': {'type': 'integer', 'maximum': 999, 'minimum': 0}, 'total_cost_usd': {'type': 'number', 'minimum': 0}, 'latency_complete': {'type': 'boolean'}, 'successful_tasks': {'type': 'integer', 'maximum': 1000, 'minimum': 0}, 'cost_per_task_usd': {'type': 'number', 'minimum': 0}, 'cost_per_successful_task_usd': {'anyOf': [{'type': 'number', 'minimum': 0}, {'type': 'null'}]}, 'success_rate_interval_95_wilson': {'type': 'array', 'items': {'type': 'number', 'maximum': 1, 'minimum': 0}, 'maxItems': 2, 'minItems': 2}, 'p95_recorded_attempt_latency_per_task_ms': {'anyOf': [{'type': 'number', 'maximum': 86400000000, 'minimum': 0}, {'type': 'null'}]}}, 'additionalProperties': True}, {'type': 'null'}]}, 'decision': {'enum': ['missing_comparison', 'collect_more_data', 'quality_regression', 'latency_data_required', 'latency_regression', 'no_economic_advantage', 'candidate_for_controlled_trial'], 'type': 'string'}, 'workflow': {'type': 'string', 'pattern': '^(?!\\s)(?![\\s\\S]*\\s$)[^\\u0000-\\u001F\\u007F]+$', 'maxLength': 80, 'minLength': 1, 'description': 'Nonempty, no control characters or surrounding whitespace. The engine additionally enforces at most 80 UTF-16 code units.'}, 'candidate': {'anyOf': [{'type': 'object', 'required': ['tasks', 'attempts', 'retry_attempts', 'successful_tasks', 'total_cost_usd', 'cost_per_task_usd', 'cost_per_successful_task_usd', 'success_rate', 'success_rate_interval_95_wilson', 'p95_recorded_attempt_latency_per_task_ms', 'latency_complete'], 'properties': {'tasks': {'type': 'integer', 'maximum': 1000, 'minimum': 1}, 'attempts': {'type': 'integer', 'maximum': 1000, 'minimum': 1}, 'success_rate': {'type': 'number', 'maximum': 1, 'minimum': 0}, 'retry_attempts': {'type': 'integer', 'maximum': 999, 'minimum': 0}, 'total_cost_usd': {'type': 'number', 'minimum': 0}, 'latency_complete': {'type': 'boolean'}, 'successful_tasks': {'type': 'integer', 'maximum': 1000, 'minimum': 0}, 'cost_per_task_usd': {'type': 'number', 'minimum': 0}, 'cost_per_successful_task_usd': {'anyOf': [{'type': 'number', 'minimum': 0}, {'type': 'null'}]}, 'success_rate_interval_95_wilson': {'type': 'array', 'items': {'type': 'number', 'maximum': 1, 'minimum': 0}, 'maxItems': 2, 'minItems': 2}, 'p95_recorded_attempt_latency_per_task_ms': {'anyOf': [{'type': 'number', 'maximum': 86400000000, 'minimum': 0}, {'type': 'null'}]}}, 'additionalProperties': True}, {'type': 'null'}]}, 'comparison': {'type': 'object', 'required': ['matched_task_ids', 'baseline_only_tasks', 'candidate_only_tasks', 'same_task_set'], 'properties': {'same_task_set': {'type': 'boolean'}, 'matched_task_ids': {'type': 'integer', 'maximum': 1000, 'minimum': 0}, 'baseline_only_tasks': {'type': 'integer', 'maximum': 1000, 'minimum': 0}, 'candidate_only_tasks': {'type': 'integer', 'maximum': 1000, 'minimum': 0}}, 'additionalProperties': True}, 'opportunity_monthly': {'anyOf': [{'type': 'object', 'required': ['status', 'basis', 'currency', 'period', 'baseline_launched_tasks_per_month', 'expected_successful_tasks_per_month', 'candidate_expected_launched_tasks_per_month', 'baseline_expected_cost_usd', 'candidate_expected_cost_usd', 'potential_cost_difference_usd', 'excludes', 'conditions'], 'properties': {'basis': {'type': 'string', 'const': 'same_expected_successful_task_volume'}, 'period': {'type': 'string', 'const': 'month'}, 'status': {'type': 'string', 'const': 'conditional_projection_not_realized'}, 'currency': {'type': 'string', 'const': 'USD'}, 'excludes': {'type': 'array', 'items': {'type': 'string'}, 'minItems': 1}, 'conditions': {'type': 'array', 'items': {'type': 'string'}, 'minItems': 1}, 'baseline_expected_cost_usd': {'type': 'number', 'minimum': 0}, 'candidate_expected_cost_usd': {'type': 'number', 'minimum': 0}, 'potential_cost_difference_usd': {'type': 'number', 'minimum': 0}, 'baseline_launched_tasks_per_month': {'type': 'integer', 'maximum': 1000000, 'minimum': 1}, 'expected_successful_tasks_per_month': {'type': 'number', 'minimum': 0}, 'candidate_expected_launched_tasks_per_month': {'type': 'number', 'minimum': 0}}, 'additionalProperties': True}, {'type': 'null'}]}, 'realized_savings_usd': {'type': 'null'}, 'automatic_deployment_authorized': {'type': 'boolean', 'const': False}}, 'additionalProperties': True}, 'maxItems': 1000, 'minItems': 1}, 'methodology': {'type': 'object', 'required': ['logical_task_key', 'success', 'duplicates', 'latency', 'quality', 'uncertainty', 'missing_costs'], 'properties': {'latency': {'type': 'string'}, 'quality': {'type': 'string'}, 'success': {'type': 'string'}, 'duplicates': {'type': 'string'}, 'uncertainty': {'type': 'string'}, 'missing_costs': {'type': 'string'}, 'logical_task_key': {'type': 'array', 'items': {'type': 'string'}, 'maxItems': 3, 'minItems': 3}}, 'additionalProperties': True}, 'schema_version': {'type': 'string', 'const': '1.0.0'}, 'data_provenance': {'type': 'string', 'const': 'user_supplied_not_verified'}, 'input_attempt_records': {'type': 'integer', 'maximum': 1000, 'minimum': 1}, 'experimental_bias_control': {'type': 'string', 'const': 'unknown'}, 'automatic_deployment_authorized': {'type': 'boolean', 'const': False}}, 'additionalProperties': True}
check_agent_quality
Quality Gate
Check whether a candidate agent workflow regresses before replacing the baseline. Supply baseline and candidate attempts on matching task IDs with client-provided success labels and costs. Returns observed success rates, task-set matching, sample-size, success, latency and cost gates, plus a decision such as collect_more_data, quality_regression or candidate_for_controlled_trial. Optional thresholds use config. It evaluates the recorded labels, not the correctness of answers or future performance. Free calculation; requires an active ALPNAI agent key; no payment or automatic deployment.
Lecture seule
Idempotent
Schéma d’entrée
{'type': 'object', 'title': 'ALPNAI recorded agent attempts', 'required': ['runs'], 'properties': {'runs': {'type': 'array', 'items': {'type': 'object', 'required': ['task_id', 'workflow', 'variant', 'cost_usd', 'success'], 'properties': {'success': {'type': 'boolean'}, 'task_id': {'type': 'string', 'pattern': '^(?!\\s)(?![\\s\\S]*\\s$)[^\\u0000-\\u001F\\u007F]+$', 'maxLength': 128, 'minLength': 1, 'description': 'Nonempty, no control characters or surrounding whitespace. The engine additionally enforces at most 128 UTF-16 code units.'}, 'variant': {'enum': ['baseline', 'candidate'], 'type': 'string'}, 'cost_usd': {'type': 'number', 'maximum': 10000, 'minimum': 0, 'description': 'USD cost of this attempt. At most six decimal places; the engine verifies whole micro-USD using floating-point tolerance. No multipleOf keyword is used, to avoid rejecting valid JSON decimals.'}, 'workflow': {'type': 'string', 'pattern': '^(?!\\s)(?![\\s\\S]*\\s$)[^\\u0000-\\u001F\\u007F]+$', 'maxLength': 80, 'minLength': 1, 'description': 'Nonempty, no control characters or surrounding whitespace. The engine additionally enforces at most 80 UTF-16 code units.'}, 'latency_ms': {'type': 'number', 'maximum': 86400000, 'minimum': 0, 'description': 'Recorded attempt duration in milliseconds; decimals are accepted. Omit if not measured.'}}, 'additionalProperties': False}, 'maxItems': 1000, 'minItems': 1}, 'config': {'type': 'object', 'required': [], 'properties': {'minSamples': {'type': 'integer', 'default': 30, 'maximum': 500, 'minimum': 2}, 'monthlyTasks': {'type': 'integer', 'maximum': 1000000, 'minimum': 1, 'description': 'Baseline logical tasks launched per month; the engine requires exactly one workflow if supplied.'}, 'minSuccessRate': {'type': 'number', 'default': 0.95, 'maximum': 1, 'minimum': 0}, 'maxP95LatencyMs': {'type': 'number', 'maximum': 86400000000, 'minimum': 0}, 'maxSuccessRateDrop': {'type': 'number', 'default': 0.02, 'maximum': 1, 'minimum': 0}}, 'additionalProperties': False}}, 'description': 'Recorded attempts, not prompts or secrets. Identical rows count as separate attempts. All three analysis tools accept this same input. The runtime validator additionally enforces six-decimal micro-USD precision, the same task_id belonging to one workflow, one workflow when monthlyTasks is given, and UTF-16 length limits. HTTP body limit: 512000 UTF-8 bytes.', 'additionalProperties': False}
Schéma de sortie
{'type': 'object', 'required': ['schema_version', 'config', 'workflows'], 'properties': {'config': {'type': 'object', 'required': ['minSamples', 'minSuccessRate', 'maxSuccessRateDrop'], 'properties': {'minSamples': {'type': 'integer', 'default': 30, 'maximum': 500, 'minimum': 2}, 'monthlyTasks': {'type': 'integer', 'maximum': 1000000, 'minimum': 1, 'description': 'Baseline logical tasks launched per month; the engine requires exactly one workflow if supplied.'}, 'minSuccessRate': {'type': 'number', 'default': 0.95, 'maximum': 1, 'minimum': 0}, 'maxP95LatencyMs': {'type': 'number', 'maximum': 86400000000, 'minimum': 0}, 'maxSuccessRateDrop': {'type': 'number', 'default': 0.02, 'maximum': 1, 'minimum': 0}}, 'additionalProperties': True}, 'workflows': {'type': 'array', 'items': {'type': 'object', 'required': ['workflow', 'comparison', 'gates', 'decision', 'baseline_success', 'candidate_success', 'automatic_deployment_authorized'], 'properties': {'gates': {'type': 'object', 'required': ['both_variants', 'minimum_distinct_tasks_per_variant', 'same_task_set', 'observed_success_rate', 'recorded_latency', 'lower_cost_per_successful_task', 'experimental_bias_control'], 'properties': {'both_variants': {'enum': ['pass', 'fail', 'unknown', 'not_requested'], 'type': 'string'}, 'same_task_set': {'enum': ['pass', 'fail', 'unknown', 'not_requested'], 'type': 'string'}, 'recorded_latency': {'enum': ['pass', 'fail', 'unknown', 'not_requested'], 'type': 'string'}, 'observed_success_rate': {'enum': ['pass', 'fail', 'unknown', 'not_requested'], 'type': 'string'}, 'experimental_bias_control': {'type': 'string', 'const': 'unknown'}, 'lower_cost_per_successful_task': {'enum': ['pass', 'fail', 'unknown', 'not_requested'], 'type': 'string'}, 'minimum_distinct_tasks_per_variant': {'enum': ['pass', 'fail', 'unknown', 'not_requested'], 'type': 'string'}}, 'additionalProperties': True}, 'decision': {'enum': ['missing_comparison', 'collect_more_data', 'quality_regression', 'latency_data_required', 'latency_regression', 'no_economic_advantage', 'candidate_for_controlled_trial'], 'type': 'string'}, 'workflow': {'type': 'string', 'pattern': '^(?!\\s)(?![\\s\\S]*\\s$)[^\\u0000-\\u001F\\u007F]+$', 'maxLength': 80, 'minLength': 1, 'description': 'Nonempty, no control characters or surrounding whitespace. The engine additionally enforces at most 80 UTF-16 code units.'}, 'comparison': {'type': 'object', 'required': ['matched_task_ids', 'baseline_only_tasks', 'candidate_only_tasks', 'same_task_set'], 'properties': {'same_task_set': {'type': 'boolean'}, 'matched_task_ids': {'type': 'integer', 'maximum': 1000, 'minimum': 0}, 'baseline_only_tasks': {'type': 'integer', 'maximum': 1000, 'minimum': 0}, 'candidate_only_tasks': {'type': 'integer', 'maximum': 1000, 'minimum': 0}}, 'additionalProperties': True}, 'baseline_success': {'anyOf': [{'type': 'number', 'maximum': 1, 'minimum': 0}, {'type': 'null'}]}, 'candidate_success': {'anyOf': [{'type': 'number', 'maximum': 1, 'minimum': 0}, {'type': 'null'}]}, 'automatic_deployment_authorized': {'type': 'boolean', 'const': False}}, 'additionalProperties': True}, 'maxItems': 1000, 'minItems': 1}, 'schema_version': {'type': 'string', 'const': '1.0.0'}}, 'additionalProperties': True}
get_catalog
ALPNAI service catalog
Discover cost, latency and quality analysis APIs, their inputs, outputs, free reproducible example, authentication and current payment availability. No payment or budget debit.
Lecture seule
Idempotent
Schéma d’entrée
{'type': 'object', 'properties': {}, 'additionalProperties': False}
get_free_sample
OpenAI source sample
Read a dated primary-source sample about OpenAI. Not real-time or exhaustive.
Lecture seule
Idempotent
Schéma d’entrée
{'type': 'object', 'properties': {}, 'additionalProperties': False}
get_order
ALPNAI order status and delivery
Follow one existing commercial order owned by the authenticated agent, including after commerce is paused. May record a finalized blockchain receipt or release an expired, never-submitted reservation. Does not create an order, authorize spending or submit a payment. Pending results provide the same order ID and Retry-After; repeat get_order, never purchase again. Requires only the existing agent key.
Accès externe
Idempotent
Schéma d’entrée
{'type': 'object', 'required': ['order_id'], 'properties': {'order_id': {'type': 'string', 'pattern': '^[-A-Za-z0-9]{8,80}$', 'description': 'The original order_id returned by the purchase.'}}, 'additionalProperties': False}
purchase_changes
Change Set purchase
Get Change Set for 0.05 USDC. Default mode sandbox creates a test receipt and debits only the test budget. Explicit mode live uses x402 only when commerce, owner mandate and billing eligibility are enabled. An unpaid result contains PaymentRequired in structuredContent and content. Retry the same tool and idempotency key with the signed PaymentPayload in params._meta["x402/payment"]. Confirmed settlement is returned in result._meta["x402/payment-response"]. For a pending order, call get_order with the original order_id; do not submit another payment. REST continuation and PAYMENT-SIGNATURE headers remain supported. Never send a private key.
Destructif
Accès externe
Idempotent
Schéma d’entrée
{'type': 'object', 'required': ['idempotency_key'], 'properties': {'mode': {'enum': ['sandbox', 'live'], 'type': 'string', 'default': 'sandbox', 'description': 'Sandbox is the default. Live explicitly requests the existing authorized x402 purchase flow.'}, 'since': {'type': 'string', 'pattern': '^\\d{4}-\\d{2}-\\d{2}$'}, 'mandate_id': {'type': 'string', 'pattern': '^[-a-zA-Z0-9]{8,80}$', 'description': 'Owner-authorized commercial mandate ID. Alternatively send X-AlpNAI-Mandate in the MCP request headers.'}, 'idempotency_key': {'type': 'string', 'pattern': '^[A-Za-z0-9_-]{8,100}$'}}, 'additionalProperties': False}
purchase_evidence
Evidence Pack purchase
Get Evidence Pack for 0.25 USDC. Default mode sandbox creates a test receipt and debits only the test budget. Explicit mode live uses x402 only when commerce, owner mandate and billing eligibility are enabled. An unpaid result contains PaymentRequired in structuredContent and content. Retry the same tool and idempotency key with the signed PaymentPayload in params._meta["x402/payment"]. Confirmed settlement is returned in result._meta["x402/payment-response"]. For a pending order, call get_order with the original order_id; do not submit another payment. REST continuation and PAYMENT-SIGNATURE headers remain supported. Never send a private key.
Destructif
Accès externe
Idempotent
Schéma d’entrée
{'type': 'object', 'required': ['idempotency_key'], 'properties': {'mode': {'enum': ['sandbox', 'live'], 'type': 'string', 'default': 'sandbox', 'description': 'Sandbox is the default. Live explicitly requests the existing authorized x402 purchase flow.'}, 'since': {'type': 'string', 'pattern': '^\\d{4}-\\d{2}-\\d{2}$'}, 'mandate_id': {'type': 'string', 'pattern': '^[-a-zA-Z0-9]{8,80}$', 'description': 'Owner-authorized commercial mandate ID. Alternatively send X-AlpNAI-Mandate in the MCP request headers.'}, 'idempotency_key': {'type': 'string', 'pattern': '^[A-Za-z0-9_-]{8,100}$'}}, 'additionalProperties': False}
purchase_snapshot
Snapshot purchase
Get Snapshot for 0.01 USDC. Default mode sandbox creates a test receipt and debits only the test budget. Explicit mode live uses x402 only when commerce, owner mandate and billing eligibility are enabled. An unpaid result contains PaymentRequired in structuredContent and content. Retry the same tool and idempotency key with the signed PaymentPayload in params._meta["x402/payment"]. Confirmed settlement is returned in result._meta["x402/payment-response"]. For a pending order, call get_order with the original order_id; do not submit another payment. REST continuation and PAYMENT-SIGNATURE headers remain supported. Never send a private key.
Destructif
Accès externe
Idempotent
Schéma d’entrée
{'type': 'object', 'required': ['idempotency_key'], 'properties': {'mode': {'enum': ['sandbox', 'live'], 'type': 'string', 'default': 'sandbox', 'description': 'Sandbox is the default. Live explicitly requests the existing authorized x402 purchase flow.'}, 'since': {'type': 'string', 'pattern': '^\\d{4}-\\d{2}-\\d{2}$'}, 'mandate_id': {'type': 'string', 'pattern': '^[-a-zA-Z0-9]{8,80}$', 'description': 'Owner-authorized commercial mandate ID. Alternatively send X-AlpNAI-Mandate in the MCP request headers.'}, 'idempotency_key': {'type': 'string', 'pattern': '^[A-Za-z0-9_-]{8,100}$'}}, 'additionalProperties': False}
save_project_report
Save an agent performance report
Compute and save an aggregate performance report in the fixed Projects destination authorized by the account owner. Uses the existing free or paid report allowance. Requires an active agent key and explicit owner write permission. Reuse request_id with identical content to retry within the same permission grant. No report reading, deletion, subscription or payment authorization. Send measurements without secrets.
Idempotent
Schéma d’entrée
{'type': 'object', 'required': ['request_id', 'title', 'input'], 'properties': {'input': {'type': 'object', 'title': 'ALPNAI recorded agent attempts', 'required': ['runs'], 'properties': {'runs': {'type': 'array', 'items': {'type': 'object', 'required': ['task_id', 'workflow', 'variant', 'cost_usd', 'success'], 'properties': {'success': {'type': 'boolean'}, 'task_id': {'type': 'string', 'pattern': '^(?!\\s)(?![\\s\\S]*\\s$)[^\\u0000-\\u001F\\u007F]+$', 'maxLength': 128, 'minLength': 1, 'description': 'Nonempty, no control characters or surrounding whitespace. The engine additionally enforces at most 128 UTF-16 code units.'}, 'variant': {'enum': ['baseline', 'candidate'], 'type': 'string'}, 'cost_usd': {'type': 'number', 'maximum': 10000, 'minimum': 0, 'description': 'USD cost of this attempt. At most six decimal places; the engine verifies whole micro-USD using floating-point tolerance. No multipleOf keyword is used, to avoid rejecting valid JSON decimals.'}, 'workflow': {'type': 'string', 'pattern': '^(?!\\s)(?![\\s\\S]*\\s$)[^\\u0000-\\u001F\\u007F]+$', 'maxLength': 80, 'minLength': 1, 'description': 'Nonempty, no control characters or surrounding whitespace. The engine additionally enforces at most 80 UTF-16 code units.'}, 'latency_ms': {'type': 'number', 'maximum': 86400000, 'minimum': 0, 'description': 'Recorded attempt duration in milliseconds; decimals are accepted. Omit if not measured.'}}, 'additionalProperties': False}, 'maxItems': 1000, 'minItems': 1}, 'config': {'type': 'object', 'required': [], 'properties': {'minSamples': {'type': 'integer', 'default': 30, 'maximum': 500, 'minimum': 2}, 'monthlyTasks': {'type': 'integer', 'maximum': 1000000, 'minimum': 1, 'description': 'Baseline logical tasks launched per month; the engine requires exactly one workflow if supplied.'}, 'minSuccessRate': {'type': 'number', 'default': 0.95, 'maximum': 1, 'minimum': 0}, 'maxP95LatencyMs': {'type': 'number', 'maximum': 86400000000, 'minimum': 0}, 'maxSuccessRateDrop': {'type': 'number', 'default': 0.02, 'maximum': 1, 'minimum': 0}}, 'additionalProperties': False}}, 'description': 'Recorded attempts, not prompts or secrets. Identical rows count as separate attempts. All three analysis tools accept this same input. The runtime validator additionally enforces six-decimal micro-USD precision, the same task_id belonging to one workflow, one workflow when monthlyTasks is given, and UTF-16 length limits. HTTP body limit: 512000 UTF-8 bytes.', 'additionalProperties': False}, 'title': {'type': 'string', 'maxLength': 80, 'minLength': 1}, 'request_id': {'type': 'string', 'pattern': '^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{12}$'}}, 'additionalProperties': False}
Historique
Modifications récentes des outils
Ajouté
purchase_evidence
17 September 2026 12:41
Ajouté
purchase_changes
17 September 2026 12:41
Ajouté
purchase_snapshot
17 September 2026 12:41
Ajouté
get_order
17 September 2026 12:41
Ajouté
save_project_report
17 September 2026 12:41
Ajouté
check_agent_quality
17 September 2026 12:41
Ajouté
analyze_agent_latency
17 September 2026 12:41
Ajouté
audit_agent_costs
17 September 2026 12:41
Ajouté
get_free_sample
17 September 2026 12:41
Ajouté
get_catalog
17 September 2026 12:41
Découvrir
Serveurs MCP similaires
Huggingface
io.github.pipeworx-io/huggingface
Provides access to Hugging Face model, dataset, and Space metadata, alongside broader structured research and data-routing tools.
Ai Model Experiments
io.github.pipeworx-io/ai-model-experiments
Runs prompts across multiple AI models and compares their outputs, costs, latency, token usage, and errors.
branchly
io.branchly/branchly
Manages an AI application’s knowledge-base content, prompts, tools, data sources, retrieval context, sessions, and usage analytic…
myriade
ai.myriade/myriade
Lets users explore and query a data warehouse through an AI data analyst agent.
HuggingFace New Dataset Release Tracker (hfdatasets)
com.a2awire/data-new-dataset-release-huggingface-tracker
Tracks newly released Hugging Face datasets, including updates, tags, and licenses.
aifu Agent Market
asia.aifu/agent-market
Discovers and hires agents or approved humans through a job ledger, verifies agent performance, and provides crypto, equity, sect…
SigRank — AI Operator Benchmarking
com.signalaf/sigrank
Benchmarks AI operator token usage, calculates cascade metrics, compares leaderboard performance, diagnoses efficiency, and simul…
Automan
io.github.i02202/automan
Offers paid AI skills for data analysis, classification, extraction, reporting, summarization, translation, text editing, copywri…