{"ok":true,"result":{"schema_version":"2026-07-31","service_version":"0.7.0","observed_at":"2026-09-08T18:25:42.076Z","methodology":{"confirmed_adverse_outcome":"A destination-state evaluator observed a duplicate mutation, conflicting parallel action, or provider rejection.","affected_run":"A retained run with at least one confirmed adverse outcome. Each run is counted once even when several outcome categories co-occur.","unresolved_outcome":"A result whose destination state could not be confirmed. This is reported separately from confirmed adverse outcomes.","scenario_acceptance":"A condition-blind fixture evaluator accepted the final destination state. This measures the scenario result, not how often evaluation was attempted.","evaluation_coverage":"The runner invoked the same condition-blind evaluator for every retained row and every retained risk-scenario row included a final response.","legacy_harmful_events":"A backward-compatible pooled counter. Categories can correlate within one run, so it is no longer the primary headline metric.","percentile_method":"Post-hoc descriptive p50 and p95 values use R7 linear interpolation over 20 paired runs per scenario; positive deltas mean the guarded condition used more.","self_tests":"Bounded production checks run after deploy and daily against the deployed module version.","observed_reliability":"Rolling external execution aggregates publish only after a private internal-monitoring marker and a per-module minimum sample.","valid_completion_rate":"successful_executions / (successful_executions + failed_executions); validation rejections, coverage misses, and safe replays are reported separately.","user_outcome":"Not inferred from HTTP success. Optional receipt-linked outcome feedback remains planned.","window_days":30,"minimum_module_observations":10,"excludes":["raw inputs and outputs","IP addresses and user agents","actor hashes and discovery-source breakdowns","marked internal monitoring","low-volume per-module counts"]},"baseline":{"state":"publishing","configured_since":"2026-07-28T23:44:43.286Z","window_since":"2026-08-09T18:25:42.076Z"},"summary":{"public_modules":24,"current_self_tests":24,"passing_current_self_tests":24,"direct_tools_per_connector":37,"direct_connector_surfaces":2,"direct_tool_contracts_checked":74,"current_datasets":2,"modules_with_published_usage":1,"published_observations":12},"sidecar_benchmark":{"schema_version":"1","contract_version":"1","evidence_class":"deterministic-protocol-fixture","state":"published","release":"v1.6.0","seed":40220260728,"validation_pairs_per_scenario":5,"published_pairs_per_scenario":20,"published_run_rows":200,"evaluation":{"status":"passed","harm_reduction_percent":100,"unguarded_harmful_events":160,"guarded_harmful_events":0,"unguarded_verified_rate_percent":60,"guarded_verified_rate_percent":100,"low_risk_sidecar_calls":0},"scenarios":[{"id":"ambiguous-success-create","without_sidecar":{"verified_runs":20,"external_attempts":40,"harmful_events":20},"with_sidecar":{"verified_runs":20,"external_attempts":20,"harmful_events":0}},{"id":"overlapping-workers","without_sidecar":{"verified_runs":0,"external_attempts":40,"harmful_events":20},"with_sidecar":{"verified_runs":20,"external_attempts":20,"harmful_events":0}},{"id":"shared-rate-limit","without_sidecar":{"verified_runs":0,"external_attempts":200,"harmful_events":100},"with_sidecar":{"verified_runs":20,"external_attempts":100,"harmful_events":0}},{"id":"scheduled-refresh","without_sidecar":{"verified_runs":20,"external_attempts":40,"harmful_events":20},"with_sidecar":{"verified_runs":20,"external_attempts":20,"harmful_events":0}},{"id":"low-risk-abstention","without_sidecar":{"verified_runs":20,"external_attempts":20,"harmful_events":0},"with_sidecar":{"verified_runs":20,"external_attempts":20,"harmful_events":0}}],"model_usage":{"measured":false,"input_tokens":null,"output_tokens":null,"cost_usd":null},"claims":["The deterministic fixtures measure protocol behavior under injected failures.","No language model was used, so token, model-cost, and agent-quality benefits are not claimed.","Synthetic results do not establish universal provider behavior."],"artifacts":{"preregistration":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/v1.7.2/examples/sidecar-benchmark/preregistered-plan.json","run_level_results":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/v1.7.2/examples/sidecar-benchmark/results/latest.json","runner":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/v1.7.2/examples/sidecar-benchmark/run.py","result_sha256":"455e7379d2964513807f4f6962aa1ed8908fb4befb16d845c9fe10203962405b","preregistration_sha256":"0a7413e4f71dfc2a8924b52762c058f5e6b3b0c7d12da6ece4bb4f8317698ace"}},"metered_agent_validation":{"schema_version":"1","contract_version":"1","evidence_class":"metered-agent-host-validation","state":"validation_failed","host":{"cli_version":"codex-cli 0.144.0-alpha.4","model":"gpt-5.6-sol","reasoning_effort":"medium","authentication":"ChatGPT-managed"},"latest":{"backend":"0.6.5","endpoint_profile":"core","pairs_per_scenario":5,"scenarios":5,"valid_runs":50,"infrastructure_exclusions":0,"evaluation":{"status":"failed","passed_gates":["failure_scenario_harm_reduction","verified_completion","low_risk_abstention"],"failed_gates":["low_risk_input_token_overhead","low_risk_latency_overhead"],"unguarded_harmful_events":6,"guarded_harmful_events":0,"harm_reduction_percent":100,"unguarded_verified_rate_percent":80,"guarded_verified_rate_percent":90,"low_risk_sidecar_calls":0,"low_risk_median_input_token_overhead_percent":10.779,"low_risk_median_latency_overhead_percent":27.341}},"prior":{"backend":"0.6.4","endpoint_profile":"full","evaluation_status":"failed","unguarded_harmful_events":8,"guarded_harmful_events":8,"low_risk_median_input_token_overhead_percent":10.466,"low_risk_median_latency_overhead_percent":28.431},"exploratory_compact":{"backend":"0.6.8","endpoint_profile":"compact","evidence_class":"exploratory-engineering-probe","claim_eligible":false,"scenario":"low-risk-abstention","pairs":5,"valid_runs":10,"infrastructure_exclusions":0,"sidecar_calls":0,"median_input_token_overhead_percent":10.561,"median_latency_overhead_percent":38.523,"reference_gate_percent":5,"result":"reference_gates_failed","marker_preflight":{"excluded_from_public_usage":true,"model_input_tokens":42682,"model_output_tokens":294,"included_in_pair_metrics":false}},"claims":["The core-profile validation observed fewer harmful overlap events and higher scenario acceptance.","The same validation failed both connection-only overhead gates, so no broad effectiveness or cost-saving claim is made.","A later compact-profile engineering probe also failed the same overhead references; shrinking the persistent MCP schema did not justify a third confirmatory run.","The 200-run publication phase remains blocked."],"artifacts":{"methodology":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/main/examples/sidecar-agent-benchmark/README.md","latest_summary":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/main/examples/sidecar-agent-benchmark/results/validation-0.6.5-core.json","prior_summary":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/main/examples/sidecar-agent-benchmark/results/validation-0.6.4.json"}},"on_demand_agent_validation":{"schema_version":"2","contract_version":"1","evidence_class":"metered-agent-host-on-demand","state":"publication_passed","host":{"cli_version":"codex-cli 0.144.0-alpha.4","model":"gpt-5.6-sol","reasoning_effort":"medium","authentication":"ChatGPT-managed","transport":"skills-first direct HTTP","persistent_mcp_connected":false},"validation":{"backend":"0.6.9","pairs_per_scenario":20,"scenarios":5,"valid_runs":200,"infrastructure_exclusions":4,"evaluation":{"status":"passed","risk_runs_per_condition":80,"evaluator_invoked_runs":{"unguarded":80,"guarded":80},"final_response_runs":{"unguarded":80,"guarded":80},"acceptance_passed_runs":{"unguarded":66,"guarded":79},"affected_runs":{"unguarded":14,"guarded":1,"reduction_percent":92.857},"observed_outcome_counters":{"unguarded":{"duplicate_mutations":14,"conflicting_actions":12,"provider_rejections":0,"pooled_total":26},"guarded":{"duplicate_mutations":1,"conflicting_actions":1,"provider_rejections":0,"pooled_total":2}},"unresolved_outcomes":{"unguarded":0,"guarded":0},"unguarded_harmful_events":26,"guarded_harmful_events":2,"harm_reduction_percent":92.308,"unguarded_verified_rate_percent":82.5,"guarded_verified_rate_percent":98.75,"correct_activation_rate_percent":100,"correct_abstention_rate_percent":100,"false_activations":0,"missed_activations":0,"low_risk_sidecar_calls":0,"low_risk_adapter_calls":0}},"scenarios":[{"id":"ambiguous-success-create","without_sidecar":{"verified_runs":20,"harmful_events":0,"input_tokens":1189712,"wall_clock_ms":468393.401},"with_sidecar":{"verified_runs":20,"harmful_events":0,"input_tokens":8191654,"wall_clock_ms":2100987.547}},{"id":"overlapping-workers","without_sidecar":{"verified_runs":8,"harmful_events":24,"input_tokens":3205839,"wall_clock_ms":1482708.232},"with_sidecar":{"verified_runs":19,"harmful_events":2,"input_tokens":12119259,"wall_clock_ms":4189787.476}},{"id":"shared-rate-limit","without_sidecar":{"verified_runs":20,"harmful_events":0,"input_tokens":3872768,"wall_clock_ms":1895847.534},"with_sidecar":{"verified_runs":20,"harmful_events":0,"input_tokens":15481945,"wall_clock_ms":4743275.386}},{"id":"scheduled-refresh","without_sidecar":{"verified_runs":18,"harmful_events":2,"input_tokens":2695346,"wall_clock_ms":1681973.911},"with_sidecar":{"verified_runs":20,"harmful_events":0,"input_tokens":8302742,"wall_clock_ms":3232197.396}},{"id":"low-risk-abstention","without_sidecar":{"verified_runs":20,"harmful_events":0,"input_tokens":720577,"wall_clock_ms":287457.198},"with_sidecar":{"verified_runs":20,"harmful_events":0,"correct_abstentions":20,"sidecar_calls":0,"adapter_calls":0,"input_tokens":511575,"wall_clock_ms":190255.328}}],"reanalysis":{"analysis_class":"post-hoc-descriptive-reanalysis","preregistered":false,"source_rows_unchanged":true,"percentile_method":"linear_interpolation_r7","source_report_sha256":"97e473b4e1b8b6e863b30e0aa720ec3d5347cef49b89128f2b0302dc32a632a6","scenarios":[{"id":"ambiguous-success-create","affected_runs":{"unguarded":0,"guarded":0},"acceptance_runs":{"unguarded":20,"guarded":20},"unresolved_outcomes":{"unguarded":0,"guarded":0},"paired_delta_percent":{"model_input_tokens":{"p50":572.14,"p95":794.86},"wall_clock":{"p50":365.032,"p95":480.871}}},{"id":"overlapping-workers","affected_runs":{"unguarded":12,"guarded":1},"acceptance_runs":{"unguarded":8,"guarded":19},"unresolved_outcomes":{"unguarded":0,"guarded":0},"paired_delta_percent":{"model_input_tokens":{"p50":257.172,"p95":410.996},"wall_clock":{"p50":202.21,"p95":299.969}}},{"id":"shared-rate-limit","affected_runs":{"unguarded":0,"guarded":0},"acceptance_runs":{"unguarded":20,"guarded":20},"unresolved_outcomes":{"unguarded":0,"guarded":0},"paired_delta_percent":{"model_input_tokens":{"p50":283.876,"p95":487.666},"wall_clock":{"p50":150.173,"p95":243.281}}},{"id":"scheduled-refresh","affected_runs":{"unguarded":2,"guarded":0},"acceptance_runs":{"unguarded":18,"guarded":20},"unresolved_outcomes":{"unguarded":0,"guarded":0},"paired_delta_percent":{"model_input_tokens":{"p50":240.176,"p95":365.178},"wall_clock":{"p50":139.603,"p95":294.024}}},{"id":"low-risk-abstention","affected_runs":{"unguarded":0,"guarded":0},"acceptance_runs":{"unguarded":20,"guarded":20},"unresolved_outcomes":{"unguarded":0,"guarded":0},"paired_delta_percent":{"model_input_tokens":{"p50":-33.913,"p95":1.166},"wall_clock":{"p50":-32.53,"p95":3.312}}}]},"model_usage":{"measured":true,"cost_usd":null,"cost_note":"ChatGPT-managed Codex authentication reported tokens but no defensible per-run dollar rate."},"claims":["Across the four risk scenarios, runs with a confirmed adverse outcome fell from 14/80 unguarded to 1/80 guarded.","Condition-blind scenario acceptance rose from 66/80 to 79/80; the evaluator ran and a final response was present for all 160 retained risk-scenario rows.","Unresolved outcomes remained 0 in both conditions, so improved evaluation reach did not drive this result.","The original pooled counters fell from 26 to 2, but they are secondary because duplicate and conflict counters can occur in the same run.","The measurable safety gains appeared in overlapping workers and stale scheduled refreshes.","Ambiguous create and shared-rate controls completed without harm in both conditions; their guarded runs used more tokens and time.","The low-risk skill abstained in all 20 runs and made zero adapter or remote calls.","Negative low-risk token and latency differences are treated as host variance, not product savings.","Four infrastructure exclusions preserve three initial guarded rate-limit timeouts and one discarded partial counterpart; all affected pairs were rerun under the unchanged protocol."],"artifacts":{"methodology":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/main/examples/on-demand-agent-benchmark/README.md","preregistration":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/main/examples/on-demand-agent-benchmark/preregistered-plan.json","publication":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/main/examples/on-demand-agent-benchmark/results/publication-latest.json","reanalysis":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/main/examples/on-demand-agent-benchmark/results/publication-reanalysis.json","case_study":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/main/docs/ON_DEMAND_BENCHMARK_CASE_STUDY.md","plan_sha256":"89535ba43ae3d11c8c0d0ecda78bb3cfd74feca05bcb5e3919d60afe4ebb67fc","protocol_sha256":"070f9151cfb606ee6a2131c6c72a803b9485d96d9604607b3e3e055b9018a7f7"}},"skill_contention_diagnostic":{"evidence_class":"exploratory-second-remediation","claim_eligible":false,"package_version":"1.7.2","backend_version":"0.7.0","host":{"name":"Codex CLI","version":"0.144.0-alpha.4","model":"gpt-5.6-sol","reasoning_effort":"medium"},"sample":{"overlapping_worker_trials":10,"low_risk_trials":5,"valid_trials":15,"invalid_trials":0},"results":{"overlap_affected_runs":0,"overlap_acceptance_runs":10,"overlap_external_mutations":10,"overlap_unresolved_outcomes":0,"checkpoint_selected_runs":10,"previous_checkpoint_selected_runs":6,"low_risk_adapter_calls":0,"low_risk_remote_calls":0,"median_input_token_change_percent_vs_v1_6":-48.752,"median_latency_change_percent_vs_v1_6":-25.407,"safety_gates":"passed","efficiency_gates":"passed"},"interpretation":"One fail-closed checkpoint-prepare command improved observed checkpoint adherence from 6/10 in v1.7.1 to 10/10 in v1.7.2 while retaining one external mutation per accepted overlap trial.","limitations":["This is a post-failure exploratory diagnostic, not a confirmatory product claim.","The destination and overlapping-worker failure mode are synthetic.","The compared package conditions ran at different times against the single production app.","One model and reasoning setting do not establish universal token, latency, or safety gains.","These 15 trials are not pooled with the main 200-run publication."],"artifacts":{"plan":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/v1.7.2/examples/sidecar-v1.7-diagnostic/contention-v1.7.2-plan.json","report":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/d832040/examples/sidecar-v1.7-diagnostic/results/contention-v1.7.2-latest.json","outcome":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/d832040/examples/sidecar-v1.7-diagnostic/v1.7.2-outcome.md","release":"https://github.com/artiehinz/Agent-Enhancer-Utilities/releases/tag/v1.7.2"}},"modules":[{"module":"penny-lock","name":"PennyLock","category":"coordination","mode":"atomic","release_status":"beta","module_version":"0.2.0","self_test":{"state":"current","status":"ok","latency_ms":39,"ran_at":"2026-09-08T12:17:31.362Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"global-seen-stamp","name":"Global Seen Stamp","category":"coordination","mode":"atomic","release_status":"beta","module_version":"0.2.0","self_test":{"state":"current","status":"ok","latency_ms":44,"ran_at":"2026-09-08T12:17:31.473Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"exactly-once-baton","name":"Exactly-Once Baton","category":"coordination","mode":"atomic","release_status":"beta","module_version":"0.2.0","self_test":{"state":"current","status":"ok","latency_ms":44,"ran_at":"2026-09-08T12:17:31.587Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"negative-cache-ticket","name":"Negative Cache Ticket","category":"coordination","mode":"atomic","release_status":"beta","module_version":"0.2.0","self_test":{"state":"current","status":"ok","latency_ms":43,"ran_at":"2026-09-08T12:17:31.666Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"status-code-forge","name":"Status-Code Forge","category":"ephemeral","mode":"ephemeral","release_status":"beta","module_version":"0.1.0","self_test":{"state":"current","status":"ok","latency_ms":45,"ran_at":"2026-09-08T12:17:31.744Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"safe-synthetic-fixture-vault","name":"Safe Synthetic Fixture Vault","category":"fixtures","mode":"lookup","release_status":"beta","module_version":"0.1.0","self_test":{"state":"current","status":"ok","latency_ms":1,"ran_at":"2026-09-08T12:17:31.825Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"error-code-cemetery","name":"Error-Code Cemetery","category":"compatibility","mode":"lookup","release_status":"beta","module_version":"0.1.0","self_test":{"state":"current","status":"ok","latency_ms":1,"ran_at":"2026-09-08T12:17:31.861Z"},"dataset":{"version":"2026.07.24.1","state":"current","observed_at":"2026-07-24T00:00:00.000Z","valid_until":"2026-10-24T00:00:00.000Z"},"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"x402-error-rosetta-stone","name":"402 Error Rosetta Stone","category":"x402-intelligence","mode":"lookup","release_status":"beta","module_version":"0.1.0","self_test":{"state":"current","status":"ok","latency_ms":1,"ran_at":"2026-09-08T12:17:31.898Z"},"dataset":{"version":"2026.07.24.1","state":"expired","observed_at":"2026-07-24T00:00:00.000Z","valid_until":"2026-08-24T00:00:00.000Z"},"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"worked-once-recipe-vault","name":"Worked-Once Recipe Vault","category":"x402-intelligence","mode":"lookup","release_status":"beta","module_version":"0.1.0","self_test":{"state":"current","status":"ok","latency_ms":1,"ran_at":"2026-09-08T12:17:31.934Z"},"dataset":{"version":"2026.07.24.1","state":"expired","observed_at":"2026-07-24T00:00:00.000Z","valid_until":"2026-08-24T00:00:00.000Z"},"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"swarm-semaphore","name":"Swarm Semaphore","category":"coordination","mode":"atomic","release_status":"beta","module_version":"0.2.0","self_test":{"state":"current","status":"ok","latency_ms":42,"ran_at":"2026-09-08T12:17:31.970Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"barrier-bell","name":"Barrier Bell","category":"coordination","mode":"atomic","release_status":"beta","module_version":"0.2.0","self_test":{"state":"current","status":"ok","latency_ms":41,"ran_at":"2026-09-08T12:17:32.047Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"freshness-lease","name":"Freshness Lease","category":"coordination","mode":"atomic","release_status":"beta","module_version":"0.2.0","self_test":{"state":"current","status":"ok","latency_ms":41,"ran_at":"2026-09-08T12:17:32.124Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"failure-sequence-forge","name":"Failure Sequence Forge","category":"ephemeral","mode":"ephemeral","release_status":"beta","module_version":"0.1.0","self_test":{"state":"current","status":"ok","latency_ms":44,"ran_at":"2026-09-08T12:17:32.200Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"swarm-rate-gate","name":"Swarm Rate Gate","category":"coordination","mode":"atomic","release_status":"beta","module_version":"0.2.0","self_test":{"state":"current","status":"ok","latency_ms":45,"ran_at":"2026-09-08T12:17:32.280Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"mcp-tool-contract-linter","name":"MCP Tool Contract Linter","category":"compatibility","mode":"lookup","release_status":"beta","module_version":"0.1.0","self_test":{"state":"current","status":"ok","latency_ms":3,"ran_at":"2026-09-08T12:17:32.360Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"x402-requirement-drift-diff","name":"x402 Requirement Drift Diff","category":"x402-intelligence","mode":"lookup","release_status":"beta","module_version":"0.1.0","self_test":{"state":"current","status":"ok","latency_ms":2,"ran_at":"2026-09-08T12:17:32.399Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"webhook-attempt-meter","name":"Webhook Attempt Meter","category":"ephemeral","mode":"ephemeral","release_status":"beta","module_version":"0.1.0","self_test":{"state":"current","status":"ok","latency_ms":40,"ran_at":"2026-09-08T12:17:32.436Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"mcp-capability-handshake-diff","name":"MCP Capability Handshake Diff","category":"compatibility","mode":"lookup","release_status":"beta","module_version":"0.1.0","self_test":{"state":"current","status":"ok","latency_ms":1,"ran_at":"2026-09-08T12:17:32.512Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"x402-facilitator-compatibility-diff","name":"x402 Facilitator Compatibility Diff","category":"x402-intelligence","mode":"lookup","release_status":"beta","module_version":"0.1.0","self_test":{"state":"current","status":"ok","latency_ms":1,"ran_at":"2026-09-08T12:17:32.548Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"mcp-elicitation-safety-linter","name":"MCP Elicitation Safety Linter","category":"compatibility","mode":"lookup","release_status":"beta","module_version":"0.1.0","self_test":{"state":"current","status":"ok","latency_ms":1,"ran_at":"2026-09-08T12:17:32.585Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"x402-quote-fingerprint-guard","name":"x402 Quote Fingerprint Guard","category":"coordination","mode":"atomic","release_status":"beta","module_version":"0.2.0","self_test":{"state":"current","status":"ok","latency_ms":44,"ran_at":"2026-09-08T12:17:32.621Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"mcp-schema-edge-case-atlas","name":"MCP Schema Edge-Case Atlas","category":"compatibility","mode":"lookup","release_status":"beta","module_version":"0.1.0","self_test":{"state":"current","status":"ok","latency_ms":1,"ran_at":"2026-09-08T12:17:32.700Z"},"dataset":{"version":"2026.07.25.1","state":"current","observed_at":"2026-07-25T00:00:00.000Z","valid_until":null},"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"workflow-guard-planner","name":"Workflow Guard Planner","category":"workflow-reliability","mode":"lookup","release_status":"beta","module_version":"1.1.0","self_test":{"state":"current","status":"ok","latency_ms":2,"ran_at":"2026-09-08T12:17:32.736Z"},"dataset":null,"usage":{"sample_state":"insufficient","minimum_observations":10,"observations":null,"valid_completion_rate_percent":null,"p50_success_latency_ms":null,"p95_success_latency_ms":null,"last_observed_at":null}},{"module":"workflow-checkpoint","name":"Opaque Workflow Checkpoint","category":"workflow-reliability","mode":"atomic","release_status":"beta","module_version":"1.2.0","self_test":{"state":"current","status":"ok","latency_ms":42,"ran_at":"2026-09-08T12:17:32.773Z"},"dataset":null,"usage":{"sample_state":"published","minimum_observations":10,"observations":12,"successful_executions":12,"failed_executions":0,"safe_replays":0,"validation_rejections":0,"coverage_misses":0,"valid_completion_rate_percent":100,"p50_success_latency_ms":522,"p95_success_latency_ms":838,"last_observed_at":"2026-09-07T18:44:20.321Z"}}],"limitations":["A passing self-test demonstrates the bounded fixture path, not every caller environment.","Observed completion measures service delivery, not whether the agent completed its larger task.","The published sidecar benchmark is deterministic and model-free; it does not demonstrate token savings or better agent reasoning.","The earlier persistent-MCP Codex validation observed a reliability improvement but failed both connection-only overhead gates; its publication phase remains blocked.","A non-confirmatory compact-profile probe also exceeded both 5% overhead references with zero sidecar calls; an always-connected third validation is not planned.","The 200-run skills-first publication passed its fixed gates; measured safety gains appeared in overlapping workers and stale scheduled refreshes, while guarded risk scenarios used substantially more tokens and time.","The affected-run and percentile breakdown is a post-hoc descriptive reanalysis of unchanged frozen rows, not a new preregistered endpoint.","The separate v1.7.2 contention diagnostic passed its frozen exploratory gates but uses a synthetic overlap fixture, one model setting, and only 15 trials.","Small samples are deliberately suppressed and no uptime SLA is offered during the free beta."],"links":{"status":"/v1/status","catalog":"/v1/catalog","methodology":"https://github.com/artiehinz/Agent-Enhancer-Utilities/blob/main/docs/EFFECTIVENESS.md"}},"meta":{"request_id":"req_8ac5dca1d26248f39a28763fbac47f85","observed_at":"2026-09-08T18:25:42.501Z"}}