Dependency graph
Every milestone and the dependencies between them, counted from the issues that cross them — drawn from the canonical edges, never by hand.
Every node below is a milestone of the canonical model, and every arrow says that issues of one wait on issues of the other, with how many depends_on entries cross it. Each milestone's own page draws its issues and the edges between them. The table under the diagram is the same edge list in text, for a reader who cannot see the picture, and names the issues behind every arrow.
flowchart LR
M_M000["M000<br/>15 of 15 done"]:::done
M_M001["M001<br/>4 of 4 done"]:::done
M_M002["M002<br/>2 of 4 done"]:::active
M_M003["M003<br/>13 of 17 done"]:::active
M_M004["M004<br/>19 of 45 done"]:::active
M_foundation["foundation<br/>0 of 0 done"]:::done
M_runtime_adapters["runtime-adapters<br/>0 of 6 done"]:::planned
M_execution_telemetry["execution-telemetry<br/>1 of 4 done"]:::active
M_cost_per_accepted_outcome["cost-per-accepted-outcome<br/>0 of 4 done"]:::blocked
M_routing_recommendations["routing-recommendations<br/>0 of 4 done"]:::blocked
M_shared_policy["shared-policy<br/>0 of 5 done"]:::blocked
M_fly_deployment["fly-deployment<br/>5 of 29 done"]:::active
M_web_sign_in["web-sign-in<br/>0 of 8 done"]:::planned
M_capability_graph["capability-graph<br/>11 of 34 done"]:::active
M_external_workspaces["external-workspaces<br/>2 of 12 done"]:::active
M_two_folder_integration["two-folder-integration<br/>1 of 31 done"]:::active
M_derived_data_integrity["derived-data-integrity<br/>1 of 3 done"]:::active
M_work_graph_github["work-graph-github<br/>10 of 12 done"]:::active
M_intent_lifecycle["intent-lifecycle<br/>0 of 3 done"]:::planned
M_intent_planning["intent-planning<br/>0 of 2 done"]:::blocked
M_development_runtime["development-runtime<br/>58 of 92 done"]:::active
M_test_evidence["test-evidence<br/>19 of 32 done"]:::active
M_public_proof["public-proof<br/>14 of 26 done"]:::active
M_adopter_upgrades["adopter-upgrades<br/>14 of 20 done"]:::active
M_a_second_adopter_is_served["a-second-adopter-is-served<br/>4 of 5 done"]:::active
M_intent_bound_work["intent-bound-work<br/>0 of 5 done"]:::blocked
M_ci_critical_path["ci-critical-path<br/>4 of 32 done"]:::active
M_measurements_are_tracked_over_time["measurements-are-tracked-over-time<br/>0 of 6 done"]:::planned
M_automation["automation<br/>3 of 6 done"]:::active
M_a_session_leaves_what_it_observed["a-session-leaves-what-it-observed<br/>3 of 3 done"]:::done
M_runtime_trust["runtime-trust<br/>6 of 10 done"]:::active
M_runtime_lifecycle["runtime-lifecycle<br/>7 of 10 done"]:::active
M_mesh_journal_converges["mesh-journal-converges<br/>4 of 14 done"]:::active
M_mcp_transports_are_reliable["mcp-transports-are-reliable<br/>2 of 7 done"]:::active
M_surfaces_never_disagree["surfaces-never-disagree<br/>8 of 18 done"]:::active
M_failure_is_visible["failure-is-visible<br/>5 of 6 done"]:::active
M_architecture_is_enforced["architecture-is-enforced<br/>6 of 8 done"]:::active
M_safe_self_update["safe-self-update<br/>4 of 12 done"]:::active
M_every_session_joins_the_mesh["every-session-joins-the-mesh<br/>3 of 5 done"]:::active
M_truthful_plan_surfaces["truthful-plan-surfaces<br/>16 of 44 done"]:::active
M_browser_compute["browser-compute<br/>0 of 1 done"]:::planned
M_idd_workspace["idd-workspace<br/>6 of 6 done"]:::done
M_idd_reads_at_a_glance["idd-reads-at-a-glance<br/>6 of 25 done"]:::active
M_a_test_fails_when_the_code_is_wrong["a-test-fails-when-the-code-is-wrong<br/>0 of 3 done"]:::planned
M_a_claim_stands_only_on_its_evidence["a-claim-stands-only-on-its-evidence<br/>0 of 20 done"]:::planned
M_a_stranger_adopts_it_unaided["a-stranger-adopts-it-unaided<br/>0 of 5 done"]:::blocked
M_the_product_shows_what_it_caught["the-product-shows-what-it-caught<br/>0 of 4 done"]:::blocked
M_prismatic_lumen["prismatic-lumen<br/>0 of 4 done"]:::planned
M_the_session_domain_is_written_once["the-session-domain-is-written-once<br/>0 of 9 done"]:::planned
M_entry_proves_what_it_handed["entry-proves-what-it-handed<br/>0 of 7 done"]:::blocked
M_knowledge_acts_on_what_it_learns["knowledge-acts-on-what-it-learns<br/>0 of 5 done"]:::planned
M_campaigns_are_read_as_provenance["campaigns-are-read-as-provenance<br/>0 of 4 done"]:::planned
M_repository_automation_is_typed["repository-automation-is-typed<br/>0 of 8 done"]:::planned
M_the_runtime_supervises_itself["the-runtime-supervises-itself<br/>0 of 4 done"]:::blocked
M_the_host_knows_its_runtimes["the-host-knows-its-runtimes<br/>0 of 2 done"]:::blocked
M_mesh_connectivity["mesh-connectivity<br/>0 of 6 done"]:::blocked
M_the_mesh_is_operated_from_one_page["the-mesh-is-operated-from-one-page<br/>0 of 2 done"]:::blocked
M_the_hub_is_reached_from_any_device["the-hub-is-reached-from-any-device<br/>0 of 2 done"]:::blocked
M_mesh_peers_execute_assigned_work["mesh-peers-execute-assigned-work<br/>0 of 2 done"]:::blocked
M_an_authorised_node_connects_from_anywhere["an-authorised-node-connects-from-anywhere<br/>1 of 5 done"]:::active
M_the_mesh_is_operated_from_any_device["the-mesh-is-operated-from-any-device<br/>0 of 1 done"]:::blocked
M_delivery_flow_is_measured["delivery-flow-is-measured<br/>0 of 6 done"]:::blocked
M_a_release_is_proven_end_to_end["a-release-is-proven-end-to-end<br/>0 of 3 done"]:::blocked
M_builds_reuse_the_mesh_cache["builds-reuse-the-mesh-cache<br/>0 of 6 done"]:::blocked
M_finished_work_reaches_a_verified_release["finished-work-reaches-a-verified-release<br/>0 of 2 done"]:::blocked
M_ordinary_commands_never_build["ordinary-commands-never-build<br/>0 of 1 done"]:::planned
M_the_ui_works_on_a_phone["the-ui-works-on-a-phone<br/>0 of 4 done"]:::planned
M_every_surface_follows_what_changed["every-surface-follows-what-changed<br/>0 of 8 done"]:::planned
M_a_visitor_understands_and_checks_majordomus["a-visitor-understands-and-checks-majordomus<br/>6 of 7 done"]:::active
M_a_derive_answers_within_a_second["a-derive-answers-within-a-second<br/>0 of 2 done"]:::blocked
M_the_plan_says_what_a_fix_unblocks["the-plan-says-what-a-fix-unblocks<br/>0 of 4 done"]:::planned
M_intent_intake["intent-intake<br/>0 of 4 done"]:::blocked
M_a_promoted_record_opens_its_message["a-promoted-record-opens-its-message<br/>0 of 5 done"]:::planned
M_a_workspace_reaches_the_repository["a-workspace-reaches-the-repository<br/>0 of 2 done"]:::blocked
M_a_provider_carries_the_contract_natively["a-provider-carries-the-contract-natively<br/>0 of 11 done"]:::planned
M_a_change_is_judged_in_minutes["a-change-is-judged-in-minutes<br/>0 of 11 done"]:::planned
M_operation_convergence["operation-convergence<br/>0 of 1 done"]:::blocked
M_what_a_workspace_decides_reaches_the_repository_with_its_source["what-a-workspace-decides-reaches-the-repository-with-its-source<br/>0 of 2 done"]:::blocked
M_every_sessions_usage_is_observed["every-sessions-usage-is-observed<br/>1 of 13 done"]:::active
M_the_ui_runs_and_explains["the-ui-runs-and-explains<br/>0 of 6 done"]:::planned
M_a_finish_reads_the_criteria_its_work_serves["a-finish-reads-the-criteria-its-work-serves<br/>0 of 1 done"]:::blocked
M_the_proof_tests_are_registered_before_they_run["the-proof-tests-are-registered-before-they-run<br/>0 of 3 done"]:::planned
M_the_proof_scorecard_is_published["the-proof-scorecard-is-published<br/>0 of 6 done"]:::planned
M_a_foreign_repository_is_adopted_and_measured["a-foreign-repository-is-adopted-and-measured<br/>0 of 7 done"]:::planned
M_the_corpus_is_registered_before_it_runs["the-corpus-is-registered-before-it-runs<br/>0 of 6 done"]:::blocked
M_a_registered_corpus_judges_majordomus["a-registered-corpus-judges-majordomus<br/>0 of 4 done"]:::planned
M_every_task_states_its_delivery_economics["every-task-states-its-delivery-economics<br/>0 of 6 done"]:::planned
M_delivery_economics_are_measured_over_declared_windows["delivery-economics-are-measured-over-declared-windows<br/>0 of 3 done"]:::blocked
M_the_mesh_is_measured_against_one_agent["the-mesh-is-measured-against-one-agent<br/>0 of 5 done"]:::blocked
M_recovery_from_injected_faults_is_proven["recovery-from-injected-faults-is-proven<br/>0 of 5 done"]:::blocked
M_a_newcomer_reaches_a_first_verified_intent["a-newcomer-reaches-a-first-verified-intent<br/>0 of 5 done"]:::planned
M_a_paste_is_decomposed_by_a_skill["a-paste-is-decomposed-by-a-skill<br/>0 of 6 done"]:::planned
M_a_prompt_reaches_the_work_it_serves["a-prompt-reaches-the-work-it-serves<br/>0 of 9 done"]:::blocked
M_an_entity_says_how_it_is_known_and_what_it_affects["an-entity-says-how-it-is-known-and-what-it-affects<br/>0 of 5 done"]:::blocked
M_knowledge_is_reconciled_before_it_is_reviewed["knowledge-is-reconciled-before-it-is-reviewed<br/>0 of 5 done"]:::blocked
M_an_inference_names_its_source_and_budget["an-inference-names-its-source-and-budget<br/>0 of 9 done"]:::blocked
M_an_outcome_becomes_a_reviewed_plan["an-outcome-becomes-a-reviewed-plan<br/>0 of 4 done"]:::planned
M_the_intent_loop_is_proved_end_to_end["the-intent-loop-is-proved-end-to-end<br/>0 of 2 done"]:::blocked
M_an_unmet_criterion_is_drafted_into_work["an-unmet-criterion-is-drafted-into-work<br/>0 of 5 done"]:::planned
M_command_and_deployment_criteria_are_judged["command-and-deployment-criteria-are-judged<br/>0 of 3 done"]:::planned
M_the_mesh_works_out_of_the_box["the-mesh-works-out-of-the-box<br/>3 of 9 done"]:::active
M_everything_is_followed_in_one_stream["everything-is-followed-in-one-stream<br/>3 of 8 done"]:::active
M_a_version_follows_the_work["a-version-follows-the-work<br/>0 of 4 done"]:::planned
M_peer_blackboard["peer-blackboard<br/>0 of 9 done"]:::planned
M_a_derive_does_only_the_work_its_inputs_changed["a-derive-does-only-the-work-its-inputs-changed<br/>1 of 12 done"]:::active
M_a_derivation_computed_once_is_reused["a-derivation-computed-once-is-reused<br/>0 of 5 done"]:::blocked
M_one_kit_on_every_surface["one-kit-on-every-surface<br/>0 of 8 done"]:::planned
M_landing_survives_any_session["landing-survives-any-session<br/>0 of 6 done"]:::planned
M_merging_and_referencing_are_enforced["merging-and-referencing-are-enforced<br/>0 of 10 done"]:::planned
M_every_model_call_is_paid_from_a_declared_source["every-model-call-is-paid-from-a-declared-source<br/>0 of 3 done"]:::planned
M_a_session_proceeds_through_the_mesh["a-session-proceeds-through-the-mesh<br/>0 of 10 done"]:::planned
M_M004 -->|1| M_a_change_is_judged_in_minutes
M_M004 -->|1| M_a_derive_answers_within_a_second
M_M004 -->|1| M_a_release_is_proven_end_to_end
M_M004 -->|1| M_finished_work_reaches_a_verified_release
M_a_claim_stands_only_on_its_evidence -->|1| M_a_stranger_adopts_it_unaided
M_a_derive_does_only_the_work_its_inputs_changed -->|5| M_a_change_is_judged_in_minutes
M_a_derive_does_only_the_work_its_inputs_changed -->|4| M_a_derivation_computed_once_is_reused
M_a_derive_does_only_the_work_its_inputs_changed -->|5| M_a_derive_answers_within_a_second
M_a_finish_reads_the_criteria_its_work_serves -->|1| M_the_intent_loop_is_proved_end_to_end
M_a_foreign_repository_is_adopted_and_measured -->|1| M_a_newcomer_reaches_a_first_verified_intent
M_a_promoted_record_opens_its_message -->|1| M_a_workspace_reaches_the_repository
M_a_promoted_record_opens_its_message -->|2| M_external_workspaces
M_a_promoted_record_opens_its_message -->|4| M_intent_intake
M_a_prompt_reaches_the_work_it_serves -->|1| M_a_provider_carries_the_contract_natively
M_a_registered_corpus_judges_majordomus -->|1| M_delivery_economics_are_measured_over_declared_windows
M_a_registered_corpus_judges_majordomus -->|1| M_the_mesh_is_measured_against_one_agent
M_a_registered_corpus_judges_majordomus -->|1| M_the_proof_scorecard_is_published
M_a_stranger_adopts_it_unaided -->|1| M_a_foreign_repository_is_adopted_and_measured
M_a_version_follows_the_work -->|1| M_public_proof
M_adopter_upgrades -->|1| M_two_folder_integration
M_an_authorised_node_connects_from_anywhere -->|1| M_mesh_connectivity
M_an_entity_says_how_it_is_known_and_what_it_affects -->|2| M_a_prompt_reaches_the_work_it_serves
M_an_entity_says_how_it_is_known_and_what_it_affects -->|1| M_knowledge_is_reconciled_before_it_is_reviewed
M_an_outcome_becomes_a_reviewed_plan -->|1| M_a_prompt_reaches_the_work_it_serves
M_an_outcome_becomes_a_reviewed_plan -->|3| M_intent_intake
M_an_outcome_becomes_a_reviewed_plan -->|1| M_the_intent_loop_is_proved_end_to_end
M_an_unmet_criterion_is_drafted_into_work -->|1| M_an_outcome_becomes_a_reviewed_plan
M_an_unmet_criterion_is_drafted_into_work -->|1| M_the_intent_loop_is_proved_end_to_end
M_architecture_is_enforced -->|1| M_a_stranger_adopts_it_unaided
M_automation -->|4| M_repository_automation_is_typed
M_campaigns_are_read_as_provenance -->|3| M_intent_intake
M_capability_graph -->|1| M_an_entity_says_how_it_is_known_and_what_it_affects
M_ci_critical_path -->|4| M_a_change_is_judged_in_minutes
M_ci_critical_path -->|1| M_measurements_are_tracked_over_time
M_command_and_deployment_criteria_are_judged -->|1| M_a_finish_reads_the_criteria_its_work_serves
M_command_and_deployment_criteria_are_judged -->|1| M_operation_convergence
M_command_and_deployment_criteria_are_judged -->|1| M_the_intent_loop_is_proved_end_to_end
M_cost_per_accepted_outcome -->|1| M_every_sessions_usage_is_observed
M_cost_per_accepted_outcome -->|1| M_every_task_states_its_delivery_economics
M_cost_per_accepted_outcome -->|1| M_operation_convergence
M_cost_per_accepted_outcome -->|2| M_the_product_shows_what_it_caught
M_delivery_flow_is_measured -->|1| M_every_task_states_its_delivery_economics
M_delivery_flow_is_measured -->|1| M_finished_work_reaches_a_verified_release
M_development_runtime -->|1| M_M004
M_development_runtime -->|1| M_a_change_is_judged_in_minutes
M_development_runtime -->|3| M_a_claim_stands_only_on_its_evidence
M_development_runtime -->|4| M_builds_reuse_the_mesh_cache
M_every_model_call_is_paid_from_a_declared_source -->|1| M_every_sessions_usage_is_observed
M_every_session_joins_the_mesh -->|1| M_a_provider_carries_the_contract_natively
M_every_sessions_usage_is_observed -->|1| M_a_foreign_repository_is_adopted_and_measured
M_every_sessions_usage_is_observed -->|2| M_an_inference_names_its_source_and_budget
M_every_sessions_usage_is_observed -->|3| M_every_model_call_is_paid_from_a_declared_source
M_every_sessions_usage_is_observed -->|1| M_every_task_states_its_delivery_economics
M_every_sessions_usage_is_observed -->|1| M_execution_telemetry
M_every_sessions_usage_is_observed -->|1| M_the_mesh_is_measured_against_one_agent
M_every_surface_follows_what_changed -->|2| M_a_prompt_reaches_the_work_it_serves
M_every_surface_follows_what_changed -->|1| M_the_mesh_is_operated_from_any_device
M_every_surface_follows_what_changed -->|1| M_the_ui_runs_and_explains
M_every_task_states_its_delivery_economics -->|4| M_delivery_economics_are_measured_over_declared_windows
M_every_task_states_its_delivery_economics -->|1| M_recovery_from_injected_faults_is_proven
M_every_task_states_its_delivery_economics -->|1| M_the_proof_scorecard_is_published
M_execution_telemetry -->|4| M_every_sessions_usage_is_observed
M_external_workspaces -->|4| M_a_promoted_record_opens_its_message
M_external_workspaces -->|7| M_what_a_workspace_decides_reaches_the_repository_with_its_source
M_fly_deployment -->|2| M_an_authorised_node_connects_from_anywhere
M_idd_reads_at_a_glance -->|1| M_the_ui_runs_and_explains
M_idd_workspace -->|1| M_a_release_is_proven_end_to_end
M_idd_workspace -->|1| M_every_surface_follows_what_changed
M_idd_workspace -->|2| M_idd_reads_at_a_glance
M_idd_workspace -->|1| M_the_ui_works_on_a_phone
M_intent_bound_work -->|1| M_entry_proves_what_it_handed
M_intent_bound_work -->|1| M_the_intent_loop_is_proved_end_to_end
M_intent_lifecycle -->|1| M_intent_planning
M_knowledge_acts_on_what_it_learns -->|1| M_knowledge_is_reconciled_before_it_is_reviewed
M_landing_survives_any_session -->|2| M_merging_and_referencing_are_enforced
M_mesh_journal_converges -->|2| M_ci_critical_path
M_mesh_journal_converges -->|1| M_landing_survives_any_session
M_mesh_journal_converges -->|1| M_mesh_peers_execute_assigned_work
M_mesh_peers_execute_assigned_work -->|1| M_landing_survives_any_session
M_prismatic_lumen -->|1| M_the_mesh_is_operated_from_any_device
M_prismatic_lumen -->|2| M_the_mesh_is_operated_from_one_page
M_prismatic_lumen -->|1| M_the_ui_runs_and_explains
M_public_proof -->|1| M_a_visitor_understands_and_checks_majordomus
M_routing_recommendations -->|1| M_every_sessions_usage_is_observed
M_runtime_adapters -->|1| M_mesh_peers_execute_assigned_work
M_runtime_lifecycle -->|1| M_safe_self_update
M_runtime_trust -->|1| M_a_workspace_reaches_the_repository
M_safe_self_update -->|1| M_landing_survives_any_session
M_safe_self_update -->|1| M_the_host_knows_its_runtimes
M_safe_self_update -->|1| M_the_hub_is_reached_from_any_device
M_surfaces_never_disagree -->|1| M_one_kit_on_every_surface
M_surfaces_never_disagree -->|1| M_the_ui_runs_and_explains
M_test_evidence -->|3| M_a_claim_stands_only_on_its_evidence
M_the_corpus_is_registered_before_it_runs -->|2| M_a_registered_corpus_judges_majordomus
M_the_corpus_is_registered_before_it_runs -->|2| M_the_mesh_is_measured_against_one_agent
M_the_hub_is_reached_from_any_device -->|2| M_the_mesh_is_operated_from_one_page
M_the_mesh_works_out_of_the_box -->|1| M_an_authorised_node_connects_from_anywhere
M_the_mesh_works_out_of_the_box -->|1| M_mesh_connectivity
M_the_mesh_works_out_of_the_box -->|1| M_the_mesh_is_operated_from_one_page
M_the_product_shows_what_it_caught -->|1| M_a_registered_corpus_judges_majordomus
M_the_product_shows_what_it_caught -->|1| M_operation_convergence
M_the_product_shows_what_it_caught -->|1| M_the_corpus_is_registered_before_it_runs
M_the_proof_scorecard_is_published -->|2| M_a_registered_corpus_judges_majordomus
M_the_proof_scorecard_is_published -->|2| M_delivery_economics_are_measured_over_declared_windows
M_the_proof_scorecard_is_published -->|1| M_recovery_from_injected_faults_is_proven
M_the_proof_scorecard_is_published -->|1| M_the_mesh_is_measured_against_one_agent
M_the_proof_tests_are_registered_before_they_run -->|2| M_a_foreign_repository_is_adopted_and_measured
M_the_proof_tests_are_registered_before_they_run -->|2| M_a_newcomer_reaches_a_first_verified_intent
M_the_proof_tests_are_registered_before_they_run -->|1| M_delivery_economics_are_measured_over_declared_windows
M_the_proof_tests_are_registered_before_they_run -->|2| M_recovery_from_injected_faults_is_proven
M_the_proof_tests_are_registered_before_they_run -->|2| M_the_corpus_is_registered_before_it_runs
M_the_proof_tests_are_registered_before_they_run -->|1| M_the_mesh_is_measured_against_one_agent
M_the_proof_tests_are_registered_before_they_run -->|3| M_the_proof_scorecard_is_published
M_the_runtime_supervises_itself -->|1| M_landing_survives_any_session
M_the_runtime_supervises_itself -->|1| M_prismatic_lumen
M_the_runtime_supervises_itself -->|1| M_the_mesh_is_operated_from_one_page
M_the_session_domain_is_written_once -->|1| M_entry_proves_what_it_handed
M_truthful_plan_surfaces -->|1| M_a_claim_stands_only_on_its_evidence
M_truthful_plan_surfaces -->|1| M_a_visitor_understands_and_checks_majordomus
M_truthful_plan_surfaces -->|1| M_command_and_deployment_criteria_are_judged
M_truthful_plan_surfaces -->|1| M_delivery_flow_is_measured
M_truthful_plan_surfaces -->|3| M_merging_and_referencing_are_enforced
M_truthful_plan_surfaces -->|2| M_one_kit_on_every_surface
M_truthful_plan_surfaces -->|1| M_the_plan_says_what_a_fix_unblocks
M_web_sign_in -->|1| M_prismatic_lumen
M_web_sign_in -->|1| M_the_mesh_is_operated_from_one_page
M_what_a_workspace_decides_reaches_the_repository_with_its_source -->|2| M_a_promoted_record_opens_its_message
M_what_a_workspace_decides_reaches_the_repository_with_its_source -->|1| M_intent_intake
classDef done stroke:#16a34a,fill:#dcfce7,color:#14532d,stroke-width:2px
classDef active stroke:#2563eb,fill:#eff6ff,color:#1e3a8a,stroke-width:2px
classDef verify stroke:#7c3aed,fill:#f5f3ff,color:#4c1d95,stroke-width:2px
classDef ready stroke:#0891b2,fill:#ecfeff,color:#164e63,stroke-width:2px
classDef planned stroke:#0891b2,fill:#ecfeff,color:#164e63,stroke-width:2px
classDef blocked stroke:#b45309,fill:#fffbeb,color:#78350f,stroke-width:2px
classDef cancelled stroke:#6b7280,fill:#f9fafb,color:#374151,stroke-width:2px
classDef superseded stroke:#6b7280,fill:#f9fafb,color:#374151,stroke-width:2pxLinked from this diagram (111)
- M000 15 of 15 done
- M001 4 of 4 done
- M002 2 of 4 done
- M003 13 of 17 done
- M004 19 of 45 done
- foundation 0 of 0 done
- runtime-adapters 0 of 6 done
- execution-telemetry 1 of 4 done
- cost-per-accepted-outcome 0 of 4 done
- routing-recommendations 0 of 4 done
- shared-policy 0 of 5 done
- fly-deployment 5 of 29 done
- web-sign-in 0 of 8 done
- capability-graph 11 of 34 done
- external-workspaces 2 of 12 done
- two-folder-integration 1 of 31 done
- derived-data-integrity 1 of 3 done
- work-graph-github 10 of 12 done
- intent-lifecycle 0 of 3 done
- intent-planning 0 of 2 done
- development-runtime 58 of 92 done
- test-evidence 19 of 32 done
- public-proof 14 of 26 done
- adopter-upgrades 14 of 20 done
- a-second-adopter-is-served 4 of 5 done
- intent-bound-work 0 of 5 done
- ci-critical-path 4 of 32 done
- measurements-are-tracked-over-time 0 of 6 done
- automation 3 of 6 done
- a-session-leaves-what-it-observed 3 of 3 done
- runtime-trust 6 of 10 done
- runtime-lifecycle 7 of 10 done
- mesh-journal-converges 4 of 14 done
- mcp-transports-are-reliable 2 of 7 done
- surfaces-never-disagree 8 of 18 done
- failure-is-visible 5 of 6 done
- architecture-is-enforced 6 of 8 done
- safe-self-update 4 of 12 done
- every-session-joins-the-mesh 3 of 5 done
- truthful-plan-surfaces 16 of 44 done
- browser-compute 0 of 1 done
- idd-workspace 6 of 6 done
- idd-reads-at-a-glance 6 of 25 done
- a-test-fails-when-the-code-is-wrong 0 of 3 done
- a-claim-stands-only-on-its-evidence 0 of 20 done
- a-stranger-adopts-it-unaided 0 of 5 done
- the-product-shows-what-it-caught 0 of 4 done
- prismatic-lumen 0 of 4 done
- the-session-domain-is-written-once 0 of 9 done
- entry-proves-what-it-handed 0 of 7 done
- knowledge-acts-on-what-it-learns 0 of 5 done
- campaigns-are-read-as-provenance 0 of 4 done
- repository-automation-is-typed 0 of 8 done
- the-runtime-supervises-itself 0 of 4 done
- the-host-knows-its-runtimes 0 of 2 done
- mesh-connectivity 0 of 6 done
- the-mesh-is-operated-from-one-page 0 of 2 done
- the-hub-is-reached-from-any-device 0 of 2 done
- mesh-peers-execute-assigned-work 0 of 2 done
- an-authorised-node-connects-from-anywhere 1 of 5 done
- the-mesh-is-operated-from-any-device 0 of 1 done
- delivery-flow-is-measured 0 of 6 done
- a-release-is-proven-end-to-end 0 of 3 done
- builds-reuse-the-mesh-cache 0 of 6 done
- finished-work-reaches-a-verified-release 0 of 2 done
- ordinary-commands-never-build 0 of 1 done
- the-ui-works-on-a-phone 0 of 4 done
- every-surface-follows-what-changed 0 of 8 done
- a-visitor-understands-and-checks-majordomus 6 of 7 done
- a-derive-answers-within-a-second 0 of 2 done
- the-plan-says-what-a-fix-unblocks 0 of 4 done
- intent-intake 0 of 4 done
- a-promoted-record-opens-its-message 0 of 5 done
- a-workspace-reaches-the-repository 0 of 2 done
- a-provider-carries-the-contract-natively 0 of 11 done
- a-change-is-judged-in-minutes 0 of 11 done
- operation-convergence 0 of 1 done
- what-a-workspace-decides-reaches-the-repository-with-its-source 0 of 2 done
- every-sessions-usage-is-observed 1 of 13 done
- the-ui-runs-and-explains 0 of 6 done
- a-finish-reads-the-criteria-its-work-serves 0 of 1 done
- the-proof-tests-are-registered-before-they-run 0 of 3 done
- the-proof-scorecard-is-published 0 of 6 done
- a-foreign-repository-is-adopted-and-measured 0 of 7 done
- the-corpus-is-registered-before-it-runs 0 of 6 done
- a-registered-corpus-judges-majordomus 0 of 4 done
- every-task-states-its-delivery-economics 0 of 6 done
- delivery-economics-are-measured-over-declared-windows 0 of 3 done
- the-mesh-is-measured-against-one-agent 0 of 5 done
- recovery-from-injected-faults-is-proven 0 of 5 done
- a-newcomer-reaches-a-first-verified-intent 0 of 5 done
- a-paste-is-decomposed-by-a-skill 0 of 6 done
- a-prompt-reaches-the-work-it-serves 0 of 9 done
- an-entity-says-how-it-is-known-and-what-it-affects 0 of 5 done
- knowledge-is-reconciled-before-it-is-reviewed 0 of 5 done
- an-inference-names-its-source-and-budget 0 of 9 done
- an-outcome-becomes-a-reviewed-plan 0 of 4 done
- the-intent-loop-is-proved-end-to-end 0 of 2 done
- an-unmet-criterion-is-drafted-into-work 0 of 5 done
- command-and-deployment-criteria-are-judged 0 of 3 done
- the-mesh-works-out-of-the-box 3 of 9 done
- everything-is-followed-in-one-stream 3 of 8 done
- a-version-follows-the-work 0 of 4 done
- peer-blackboard 0 of 9 done
- a-derive-does-only-the-work-its-inputs-changed 1 of 12 done
- a-derivation-computed-once-is-reused 0 of 5 done
- one-kit-on-every-surface 0 of 8 done
- landing-survives-any-session 0 of 6 done
- merging-and-referencing-are-enforced 0 of 10 done
- every-model-call-is-paid-from-a-declared-source 0 of 3 done
- a-session-proceeds-through-the-mesh 0 of 10 done
Edges
128 edge(s) between 111 milestones, drawn from 908 issue dependencies. Every one of those is in site/data/generated/plan.json, and majordomus plan graph draws them all.
| must finish | before | issue dependencies |
|---|---|---|
| M004 Performance is executable evidence, and the hot path does no canonical work twice | a-change-is-judged-in-minutes A change is derived in under a second and judged in minutes, and nothing that could have changed is skipped | 1: I0832 → I3235 |
| M004 Performance is executable evidence, and the hot path does no canonical work twice | a-derive-answers-within-a-second An ordinary derive answers in under a second, runs independent nodes in parallel, and a budget gate holds it | 1: I2068 → I3236 |
| M004 Performance is executable evidence, and the hot path does no canonical work twice | a-release-is-proven-end-to-end Whether a commit may be released, and whether its release arrived, is proved by one read and one case | 1: I2717 → I3213 |
| M004 Performance is executable evidence, and the hot path does no canonical work twice | finished-work-reaches-a-verified-release Finished work reaches a verified release at the pace it is finished | 1: I2709 → I3200 |
| a-claim-stands-only-on-its-evidence A claim that something is enforced, tested or delivered stands only on recorded evidence that would have caught it being false | a-stranger-adopts-it-unaided A stranger installs Majordomus and is stopped from a false done in ten minutes, and every outside first run teaches the tool | 1: I3123 → I3128 |
| a-derive-does-only-the-work-its-inputs-changed A derive does only the work its inputs changed, and can say why | a-change-is-judged-in-minutes A change is derived in under a second and judged in minutes, and nothing that could have changed is skipped | 5: I2521 → I3233 I2521 → I3234 I2521 → I3235 I2525 → I3234 I2527 → I3238 |
| a-derive-does-only-the-work-its-inputs-changed A derive does only the work its inputs changed, and can say why | a-derivation-computed-once-is-reused A derivation computed once is reused by every worktree and by CI, and the incremental path becomes the default on evidence | 4: I2525 → I2529 I2527 → I2529 I2527 → I2532 I2527 → I2533 |
| a-derive-does-only-the-work-its-inputs-changed A derive does only the work its inputs changed, and can say why | a-derive-answers-within-a-second An ordinary derive answers in under a second, runs independent nodes in parallel, and a budget gate holds it | 5: I2521 → I3236 I2521 → I3237 I2522 → I3237 I2525 → I3236 I2527 → I3236 |
| a-finish-reads-the-criteria-its-work-serves A finish reads what its work serves | the-intent-loop-is-proved-end-to-end The intent loop is proved end to end, and the plan and the documents say what it does | 1: I3259 → I3261 |
| a-foreign-repository-is-adopted-and-measured A repository not built around Majordomus is adopted from the published installer to an accepted finish, and its overhead and friction are recorded without exposing what its basis does not permit | a-newcomer-reaches-a-first-verified-intent A newcomer reaches an intent whose verdict is satisfied, in a repository of their own, using only what the getting-started page shows | 1: I3361 → I3397 |
| a-promoted-record-opens-its-message Every record promoted from a workspace cites the message it came from, and the UI opens it for the operator who holds it | a-workspace-reaches-the-repository A vendor workspace reads the repository, and proposes to it only where the owner has decided it may | 1: I3281 → I3286 |
| a-promoted-record-opens-its-message Every record promoted from a workspace cites the message it came from, and the UI opens it for the operator who holds it | external-workspaces An external workspace is read the way this repository reads anything else, and its content never becomes public by accident | 2: I3281 → I3173 I3284 → I3173 |
| a-promoted-record-opens-its-message Every record promoted from a workspace cites the message it came from, and the UI opens it for the operator who holds it | intent-intake A received brief is a campaign revision whose requirements are reconciled into the one plan | 4: I3280 → I3273 I3280 → I3275 I3281 → I3273 I3283 → I3275 |
| a-prompt-reaches-the-work-it-serves A captured prompt is matched to its existing intent and recorded on the task that follows, joined to its outcome and evidence, shown on /ui and the Cockpit, read only on its own machine and never published | a-provider-carries-the-contract-natively Each agent client carries the contract through its own extension points, generated once and checked for drift | 1: I3406 → I3410 |
| a-registered-corpus-judges-majordomus The registered corpus judges the same agent with and without Majordomus, by an exact task-level test on accepted results, and the verdict is published with its losses | delivery-economics-are-measured-over-declared-windows Delivery economics are measured on corpus runs and on real work over a window registered before it opens, and published with their denominators and losses | 1: I3373 → I3383 |
| a-registered-corpus-judges-majordomus The registered corpus judges the same agent with and without Majordomus, by an exact task-level test on accepted results, and the verdict is published with its losses | the-mesh-is-measured-against-one-agent Coordinated workers on the mesh are compared with a single agent on the same batch tasks, at quality held by a registered non-inferiority guard | 1: I3371 → I3385 |
| a-registered-corpus-judges-majordomus The registered corpus judges the same agent with and without Majordomus, by an exact task-level test on accepted results, and the verdict is published with its losses | the-proof-scorecard-is-published One derived scorecard publishes every proof test's computed status, losses included, on every surface, and no value claim stands without a passed test | 1: I3366 → I3354 |
| a-stranger-adopts-it-unaided A stranger installs Majordomus and is stopped from a false done in ten minutes, and every outside first run teaches the tool | a-foreign-repository-is-adopted-and-measured A repository not built around Majordomus is adopted from the published installer to an accepted finish, and its overhead and friction are recorded without exposing what its basis does not permit | 1: I3127 → I3361 |
| a-version-follows-the-work Landed work is released as a measured version, and how long it has waited is a finding | public-proof A visitor sees the product prove itself, and can check every claim it makes | 1: I3076 → I3085 |
| adopter-upgrades An adopting repository survives a Majordomus upgrade, and what the tool vendors into it is adopter-neutral | two-folder-integration A repository that adopts Majordomus gains two directories and two lines, and can give them back | 1: I1980 → I2873 |
| an-authorised-node-connects-from-anywhere An authorised node reaches another across LAN, tailnet or NAT, and says how | mesh-connectivity Trusted nodes connect across NAT through proven transports, and every surface says which path | 1: I3176 → I3188 |
| an-entity-says-how-it-is-known-and-what-it-affects One entity read names how its record says it is known and when it was observed, identically on every surface, and what a change affects is derived from the same index | a-prompt-reaches-the-work-it-serves A captured prompt is matched to its existing intent and recorded on the task that follows, joined to its outcome and evidence, shown on /ui and the Cockpit, read only on its own machine and never published | 2: I3412 → I3430 I3427 → I3430 |
| an-entity-says-how-it-is-known-and-what-it-affects One entity read names how its record says it is known and when it was observed, identically on every surface, and what a change affects is derived from the same index | knowledge-is-reconciled-before-it-is-reviewed What a session proposes as knowledge is reconciled with what the repository already knows before a person reviews it, and a promotion stands on evidence that resolves | 1: I3412 → I3414 |
| an-outcome-becomes-a-reviewed-plan A desired outcome becomes a draft plan that is checked before a record is written | a-prompt-reaches-the-work-it-serves A captured prompt is matched to its existing intent and recorded on the task that follows, joined to its outcome and evidence, shown on /ui and the Cockpit, read only on its own machine and never published | 1: I3251 → I3408 |
| an-outcome-becomes-a-reviewed-plan A desired outcome becomes a draft plan that is checked before a record is written | intent-intake A received brief is a campaign revision whose requirements are reconciled into the one plan | 3: I3251 → I3273 I3252 → I3273 I3254 → I3273 |
| an-outcome-becomes-a-reviewed-plan A desired outcome becomes a draft plan that is checked before a record is written | the-intent-loop-is-proved-end-to-end The intent loop is proved end to end, and the plan and the documents say what it does | 1: I3252 → I3260 |
| an-unmet-criterion-is-drafted-into-work An unmet criterion is drafted into work from the evidence, and the draft is the same on every run | an-outcome-becomes-a-reviewed-plan A desired outcome becomes a draft plan that is checked before a record is written | 1: I3247 → I3252 |
| an-unmet-criterion-is-drafted-into-work An unmet criterion is drafted into work from the evidence, and the draft is the same on every run | the-intent-loop-is-proved-end-to-end The intent loop is proved end to end, and the plan and the documents say what it does | 1: I3247 → I3260 |
| architecture-is-enforced A violated invariant fails a gate, and green means every required check ran | a-stranger-adopts-it-unaided A stranger installs Majordomus and is stopped from a false done in ten minutes, and every outside first run teaches the tool | 1: I2162 → I3127 |
| automation Repository automation is scripted capabilities over typed Rust, not shell | repository-automation-is-typed Repository automation is typed capabilities that Rhai orchestrates, and domain-logic shell only shrinks | 4: I1800 → I3306 I1801 → I3157 I1801 → I3305 I1804 → I3159 |
| campaigns-are-read-as-provenance A prompt pack is a campaign the index serves and the plan cites | intent-intake A received brief is a campaign revision whose requirements are reconciled into the one plan | 3: I3167 → I3272 I3168 → I3275 I3169 → I3272 |
| capability-graph One capability graph, two projections, and no second inventory of what this repository can do | an-entity-says-how-it-is-known-and-what-it-affects One entity read names how its record says it is known and when it was observed, identically on every surface, and what a change affects is derived from the same index | 1: I3314 → I3428 |
| ci-critical-path The critical path from a change to its verdict, and from a merge to its verified deploy, runs under a minute warm | a-change-is-judged-in-minutes A change is derived in under a second and judged in minutes, and nothing that could have changed is skipped | 4: I2348 → I3238 I2350 → I3238 I2350 → I3240 I2350 → I3242 |
| ci-critical-path The critical path from a change to its verdict, and from a merge to its verified deploy, runs under a minute warm | measurements-are-tracked-over-time Coverage and CI cost are kept, exported and drawn over time, from the one measurement each has | 1: I3919 → I3932 |
| command-and-deployment-criteria-are-judged A command or deployment criterion is met by a recorded run of exactly that command, or an observation of exactly that deployment, current at the presented revision | a-finish-reads-the-criteria-its-work-serves A finish reads what its work serves | 1: I3255 → I3259 |
| command-and-deployment-criteria-are-judged A command or deployment criterion is met by a recorded run of exactly that command, or an observation of exactly that deployment, current at the presented revision | operation-convergence After the alpha, the time from an intent to its verified satisfaction is measured from the records | 1: I3255 → I3276 |
| command-and-deployment-criteria-are-judged A command or deployment criterion is met by a recorded run of exactly that command, or an observation of exactly that deployment, current at the presented revision | the-intent-loop-is-proved-end-to-end The intent loop is proved end to end, and the plan and the documents say what it does | 1: I3255 → I3260 |
| cost-per-accepted-outcome Cost per accepted outcome, only on measured data | every-sessions-usage-is-observed What every AI session consumed is observed from the provider, attributed to the work it served, and never estimated into a fact | 1: I0402 → I3295 |
| cost-per-accepted-outcome Cost per accepted outcome, only on measured data | every-task-states-its-delivery-economics Every task states its delivery economics: wall time, typed human interventions, CI reruns, spend and acceptance, each from a named record or said not observed | 1: I0401 → I3381 |
| cost-per-accepted-outcome Cost per accepted outcome, only on measured data | operation-convergence After the alpha, the time from an intent to its verified satisfaction is measured from the records | 1: I0403 → I3276 |
| cost-per-accepted-outcome Cost per accepted outcome, only on measured data | the-product-shows-what-it-caught What finish caught is counted from the ledger and compared on the benchmark, never asserted | 2: I0401 → I3130 I0401 → I3133 |
| delivery-flow-is-measured The way from a pull request to a verified publication is measured per phase, and its bounds are read where work starts | every-task-states-its-delivery-economics Every task states its delivery economics: wall time, typed human interventions, CI reruns, spend and acceptance, each from a named record or said not observed | 1: I3202 → I3379 |
| delivery-flow-is-measured The way from a pull request to a verified publication is measured per phase, and its bounds are read where work starts | finished-work-reaches-a-verified-release Finished work reaches a verified release at the pace it is finished | 1: I3204 → I3212 |
| development-runtime The runtime has a development surface | M004 Performance is executable evidence, and the hot path does no canonical work twice | 1: I2489 → I2663 |
| development-runtime The runtime has a development surface | a-change-is-judged-in-minutes A change is derived in under a second and judged in minutes, and nothing that could have changed is skipped | 1: I2472 → I3242 |
| development-runtime The runtime has a development surface | a-claim-stands-only-on-its-evidence A claim that something is enforced, tested or delivered stands only on recorded evidence that would have caught it being false | 3: I2962 → I3117 I2963 → I3117 I2965 → I3122 |
| development-runtime The runtime has a development surface | builds-reuse-the-mesh-cache Builds reuse the mesh cache: compiled Rust, the cargo build directory, mix deps and npm packages, keyed by platform, toolchain and inputs | 4: I2670 → I3950 I2670 → I3951 I2671 → I3952 I2673 → I3953 |
| every-model-call-is-paid-from-a-declared-source How a model call Majordomus starts can be paid is declared by the vendor with evidence, and each call names the source it was declared to draw on | every-sessions-usage-is-observed What every AI session consumed is observed from the provider, attributed to the work it served, and never estimated into a fact | 1: I3660 → I3292 |
| every-session-joins-the-mesh Every LLM session joins the Majordomus mesh at start, told the mesh as work | a-provider-carries-the-contract-natively Each agent client carries the contract through its own extension points, generated once and checked for drift | 1: I2734 → I3303 |
| every-sessions-usage-is-observed What every AI session consumed is observed from the provider, attributed to the work it served, and never estimated into a fact | a-foreign-repository-is-adopted-and-measured A repository not built around Majordomus is adopted from the published installer to an accepted finish, and its overhead and friction are recorded without exposing what its basis does not permit | 1: I3299 → I3364 |
| every-sessions-usage-is-observed What every AI session consumed is observed from the provider, attributed to the work it served, and never estimated into a fact | an-inference-names-its-source-and-budget Every inference a decision rests on names its model and source, stays inside a declared budget, counts independent sources rather than answers, and is never reached from a hook path | 2: I3292 → I3422 I3295 → I3421 |
| every-sessions-usage-is-observed What every AI session consumed is observed from the provider, attributed to the work it served, and never estimated into a fact | every-model-call-is-paid-from-a-declared-source How a model call Majordomus starts can be paid is declared by the vendor with evidence, and each call names the source it was declared to draw on | 3: I3292 → I3665 I3292 → I3666 I3295 → I3667 |
| every-sessions-usage-is-observed What every AI session consumed is observed from the provider, attributed to the work it served, and never estimated into a fact | every-task-states-its-delivery-economics Every task states its delivery economics: wall time, typed human interventions, CI reruns, spend and acceptance, each from a named record or said not observed | 1: I3293 → I3380 |
| every-sessions-usage-is-observed What every AI session consumed is observed from the provider, attributed to the work it served, and never estimated into a fact | execution-telemetry Execution telemetry, only from providers that expose it honestly | 1: I3288 → I0303 |
| every-sessions-usage-is-observed What every AI session consumed is observed from the provider, attributed to the work it served, and never estimated into a fact | the-mesh-is-measured-against-one-agent Coordinated workers on the mesh are compared with a single agent on the same batch tasks, at quality held by a registered non-inferiority guard | 1: I3298 → I3386 |
| every-surface-follows-what-changed Every surface follows what changed in the repository by push, and a surface that does not says so | a-prompt-reaches-the-work-it-serves A captured prompt is matched to its existing intent and recorded on the task that follows, joined to its outcome and evidence, shown on /ui and the Cockpit, read only on its own machine and never published | 2: I3431 → I3409 I3431 → I3430 |
| every-surface-follows-what-changed Every surface follows what changed in the repository by push, and a surface that does not says so | the-mesh-is-operated-from-any-device The owner understands and operates the whole mesh from one page, on any device | 1: I3222 → I3181 |
| every-surface-follows-what-changed Every surface follows what changed in the repository by push, and a surface that does not says so | the-ui-runs-and-explains The UI runs any capability and explains what needs the owner, whose version it shows and where its time went | 1: I3221 → I3225 |
| every-task-states-its-delivery-economics Every task states its delivery economics: wall time, typed human interventions, CI reruns, spend and acceptance, each from a named record or said not observed | delivery-economics-are-measured-over-declared-windows Delivery economics are measured on corpus runs and on real work over a window registered before it opens, and published with their denominators and losses | 4: I3376 → I3382 I3377 → I3384 I3378 → I3384 I3381 → I3384 |
| every-task-states-its-delivery-economics Every task states its delivery economics: wall time, typed human interventions, CI reruns, spend and acceptance, each from a named record or said not observed | recovery-from-injected-faults-is-proven A killed worker, a cut connection and a conflict are injected on purpose, and recovery is proven without lost work or duplicate completion | 1: I3376 → I3390 |
| every-task-states-its-delivery-economics Every task states its delivery economics: wall time, typed human interventions, CI reruns, spend and acceptance, each from a named record or said not observed | the-proof-scorecard-is-published One derived scorecard publishes every proof test's computed status, losses included, on every surface, and no value claim stands without a passed test | 1: I3376 → I3355 |
| execution-telemetry Execution telemetry, only from providers that expose it honestly | every-sessions-usage-is-observed What every AI session consumed is observed from the provider, attributed to the work it served, and never estimated into a fact | 4: I0301 → I3288 I0301 → I3290 I0301 → I3292 I0304 → I3294 |
| external-workspaces An external workspace is read the way this repository reads anything else, and its content never becomes public by accident | a-promoted-record-opens-its-message Every record promoted from a workspace cites the message it came from, and the UI opens it for the operator who holds it | 4: I1301 → I3281 I1304 → I3284 I1308 → I3282 I1309 → I3283 |
| external-workspaces An external workspace is read the way this repository reads anything else, and its content never becomes public by accident | what-a-workspace-decides-reaches-the-repository-with-its-source What the owner decides in a vendor workspace reaches the repository, and every promoted record opens the message it came from | 7: I1302 → I3278 I1302 → I3279 I1303 → I3278 I1305 → I3278 I1305 → I3279 I1306 → I3278 I1308 → I3278 |
| fly-deployment The container image is a projection of the repository model, and its smoke is its evidence | an-authorised-node-connects-from-anywhere An authorised node reaches another across LAN, tailnet or NAT, and says how | 2: I3820 → I3826 I3822 → I3827 |
| idd-reads-at-a-glance The intent workspace reads at a glance and leads everywhere it names | the-ui-runs-and-explains The UI runs any capability and explains what needs the owner, whose version it shows and where its time went | 1: I2902 → I3228 |
| idd-workspace The UI is where the owner directs the work, from an intent to its next move | a-release-is-proven-end-to-end Whether a commit may be released, and whether its release arrived, is proved by one read and one case | 1: I2812 → I3213 |
| idd-workspace The UI is where the owner directs the work, from an intent to its next move | every-surface-follows-what-changed Every surface follows what changed in the repository by push, and a surface that does not says so | 1: I2783 → I3220 |
| idd-workspace The UI is where the owner directs the work, from an intent to its next move | idd-reads-at-a-glance The intent workspace reads at a glance and leads everywhere it names | 2: I2812 → I2906 I2812 → I2907 |
| idd-workspace The UI is where the owner directs the work, from an intent to its next move | the-ui-works-on-a-phone Every first-party surface works on a phone, and the primary flow is accepted there | 1: I2812 → I3216 |
| intent-bound-work A task is bound to the intent it serves | entry-proves-what-it-handed An episode records the governance and briefing it was handed, and a change is told | 1: I2003 → I3150 |
| intent-bound-work A task is bound to the intent it serves | the-intent-loop-is-proved-end-to-end The intent loop is proved end to end, and the plan and the documents say what it does | 1: I2001 → I3261 |
| intent-lifecycle Work is traceable to the intent it serves, and an intent is satisfied only by evidence | intent-planning The plan of an intent is held to it before execution | 1: I1900 → I1901 |
| knowledge-acts-on-what-it-learns The repository decides when to evolve its layer, and a session carries it out | knowledge-is-reconciled-before-it-is-reviewed What a session proposes as knowledge is reconciled with what the repository already knows before a person reviews it, and a promotion stands on evidence that resolves | 1: I3164 → I3413 |
| landing-survives-any-session Landing green work survives any session, any restart and any machine | merging-and-referencing-are-enforced Merging and referencing are enforced: every pull request carries its metadata, every reference resolves, every number is taken once | 2: I3915 → I3976 I3915 → I3978 |
| mesh-journal-converges The cooperation journal is durable, converges under any delivery order, and never lets a stale claim win | ci-critical-path The critical path from a change to its verdict, and from a merge to its verified deploy, runs under a minute warm | 2: I2874 → I3832 I2874 → I3833 |
| mesh-journal-converges The cooperation journal is durable, converges under any delivery order, and never lets a stale claim win | landing-survives-any-session Landing green work survives any session, any restart and any machine | 1: I3174 → I3912 |
| mesh-journal-converges The cooperation journal is durable, converges under any delivery order, and never lets a stale claim win | mesh-peers-execute-assigned-work Work is offered to a peer by capability under a fenced lease, and runs there as an execution | 1: I3174 → I3198 |
| mesh-peers-execute-assigned-work Work is offered to a peer by capability under a fenced lease, and runs there as an execution | landing-survives-any-session Landing green work survives any session, any restart and any machine | 1: I3198 → I3912 |
| prismatic-lumen Prismatic Lumen: the owner oversees and moves the whole mesh from one page, which owns nothing | the-mesh-is-operated-from-any-device The owner understands and operates the whole mesh from one page, on any device | 1: I3138 → I3181 |
| prismatic-lumen Prismatic Lumen: the owner oversees and moves the whole mesh from one page, which owns nothing | the-mesh-is-operated-from-one-page The owner operates the mesh from one page: every runtime readable, one action authorised end to end | 2: I3138 → I3191 I3139 → I3192 |
| prismatic-lumen Prismatic Lumen: the owner oversees and moves the whole mesh from one page, which owns nothing | the-ui-runs-and-explains The UI runs any capability and explains what needs the owner, whose version it shows and where its time went | 1: I3138 → I3230 |
| public-proof A visitor sees the product prove itself, and can check every claim it makes | a-visitor-understands-and-checks-majordomus A first-time visitor of majordomus.dev understands what Majordomus does and can check every claim it makes | 1: I2417 → I3223 |
| routing-recommendations Routing recommendations derived from measured cost | every-sessions-usage-is-observed What every AI session consumed is observed from the provider, attributed to the work it served, and never estimated into a fact | 1: I0501 → I3297 |
| runtime-adapters Profiles become runtime constraints rather than advice | mesh-peers-execute-assigned-work Work is offered to a peer by capability under a fenced lease, and runs there as an execution | 1: I0202 → I3199 |
| runtime-lifecycle One usable server per checkout starts, attaches, stops and reports its state without races or hidden kills | safe-self-update Every mesh node follows the published release by itself, versions meet safely, and a release's provenance can be checked | 1: I2127 → I2163 |
| runtime-trust A runtime is safe to reach from another machine, and joining the mesh trusts nobody by accident | a-workspace-reaches-the-repository A vendor workspace reads the repository, and proposes to it only where the owner has decided it may | 1: I2123 → I3285 |
| safe-self-update Every mesh node follows the published release by itself, versions meet safely, and a release's provenance can be checked | landing-survives-any-session Landing green work survives any session, any restart and any machine | 1: I2722 → I3913 |
| safe-self-update Every mesh node follows the published release by itself, versions meet safely, and a release's provenance can be checked | the-host-knows-its-runtimes A host knows every Majordomus runtime it runs, with no project open | 1: I3177 → I3185 |
| safe-self-update Every mesh node follows the published release by itself, versions meet safely, and a release's provenance can be checked | the-hub-is-reached-from-any-device The hub is reached privately from any of the owner's devices, and a public ingress is configuration | 1: I2722 → I3195 |
| surfaces-never-disagree Every surface projects the same state, says when it was observed, and keeps no list of its own | one-kit-on-every-surface Code, charts, controls, embeds and shell blocks are one kit on every surface, declared in the design book and held there by a gate | 1: I2654 → I3792 |
| surfaces-never-disagree Every surface projects the same state, says when it was observed, and keeps no list of its own | the-ui-runs-and-explains The UI runs any capability and explains what needs the owner, whose version it shows and where its time went | 1: I2770 → I3229 |
| test-evidence Every verifiable claim and widget resolves to revision-bound test evidence, on every surface | a-claim-stands-only-on-its-evidence A claim that something is enforced, tested or delivered stands only on recorded evidence that would have caught it being false | 3: I2786 → I3112 I2786 → I3119 I2786 → I3124 |
| the-corpus-is-registered-before-it-runs A corpus of 20 tasks, stratified by a difficulty proxy measured and registered before the new tasks are written, is registered with its analysis, run order, stopping rule and a pinned release | a-registered-corpus-judges-majordomus The registered corpus judges the same agent with and without Majordomus, by an exact task-level test on accepted results, and the verdict is published with its losses | 2: I3367 → I3371 I3370 → I3372 |
| the-corpus-is-registered-before-it-runs A corpus of 20 tasks, stratified by a difficulty proxy measured and registered before the new tasks are written, is registered with its analysis, run order, stopping rule and a pinned release | the-mesh-is-measured-against-one-agent Coordinated workers on the mesh are compared with a single agent on the same batch tasks, at quality held by a registered non-inferiority guard | 2: I3367 → I3387 I3375 → I3386 |
| the-hub-is-reached-from-any-device The hub is reached privately from any of the owner's devices, and a public ingress is configuration | the-mesh-is-operated-from-one-page The owner operates the mesh from one page: every runtime readable, one action authorised end to end | 2: I3196 → I3191 I3196 → I3192 |
| the-mesh-works-out-of-the-box The mesh works out of the box: trusted machines find each other by every means and their sessions cooperate from the first turn | an-authorised-node-connects-from-anywhere An authorised node reaches another across LAN, tailnet or NAT, and says how | 1: I2410 → I3176 |
| the-mesh-works-out-of-the-box The mesh works out of the box: trusted machines find each other by every means and their sessions cooperate from the first turn | mesh-connectivity Trusted nodes connect across NAT through proven transports, and every surface says which path | 1: I2410 → I3187 |
| the-mesh-works-out-of-the-box The mesh works out of the box: trusted machines find each other by every means and their sessions cooperate from the first turn | the-mesh-is-operated-from-one-page The owner operates the mesh from one page: every runtime readable, one action authorised end to end | 1: I2414 → I3191 |
| the-product-shows-what-it-caught What finish caught is counted from the ledger and compared on the benchmark, never asserted | a-registered-corpus-judges-majordomus The registered corpus judges the same agent with and without Majordomus, by an exact task-level test on accepted results, and the verdict is published with its losses | 1: I3133 → I3371 |
| the-product-shows-what-it-caught What finish caught is counted from the ledger and compared on the benchmark, never asserted | operation-convergence After the alpha, the time from an intent to its verified satisfaction is measured from the records | 1: I3133 → I3276 |
| the-product-shows-what-it-caught What finish caught is counted from the ledger and compared on the benchmark, never asserted | the-corpus-is-registered-before-it-runs A corpus of 20 tasks, stratified by a difficulty proxy measured and registered before the new tasks are written, is registered with its analysis, run order, stopping rule and a pinned release | 1: I3133 → I3370 |
| the-proof-scorecard-is-published One derived scorecard publishes every proof test's computed status, losses included, on every surface, and no value claim stands without a passed test | a-registered-corpus-judges-majordomus The registered corpus judges the same agent with and without Majordomus, by an exact task-level test on accepted results, and the verdict is published with its losses | 2: I3354 → I3372 I3356 → I3371 |
| the-proof-scorecard-is-published One derived scorecard publishes every proof test's computed status, losses included, on every surface, and no value claim stands without a passed test | delivery-economics-are-measured-over-declared-windows Delivery economics are measured on corpus runs and on real work over a window registered before it opens, and published with their denominators and losses | 2: I3354 → I3384 I3356 → I3384 |
| the-proof-scorecard-is-published One derived scorecard publishes every proof test's computed status, losses included, on every surface, and no value claim stands without a passed test | recovery-from-injected-faults-is-proven A killed worker, a cut connection and a conflict are injected on purpose, and recovery is proven without lost work or duplicate completion | 1: I3354 → I3394 |
| the-proof-scorecard-is-published One derived scorecard publishes every proof test's computed status, losses included, on every surface, and no value claim stands without a passed test | the-mesh-is-measured-against-one-agent Coordinated workers on the mesh are compared with a single agent on the same batch tasks, at quality held by a registered non-inferiority guard | 1: I3354 → I3389 |
| the-proof-tests-are-registered-before-they-run The six proof tests are registered as validated protocols, each frozen by its own digest, and every trial is a typed record that publishes only what its basis permits | a-foreign-repository-is-adopted-and-measured A repository not built around Majordomus is adopted from the published installer to an accepted finish, and its overhead and friction are recorded without exposing what its basis does not permit | 2: I3352 → I3360 I3353 → I3360 |
| the-proof-tests-are-registered-before-they-run The six proof tests are registered as validated protocols, each frozen by its own digest, and every trial is a typed record that publishes only what its basis permits | a-newcomer-reaches-a-first-verified-intent A newcomer reaches an intent whose verdict is satisfied, in a repository of their own, using only what the getting-started page shows | 2: I3352 → I3398 I3353 → I3398 |
| the-proof-tests-are-registered-before-they-run The six proof tests are registered as validated protocols, each frozen by its own digest, and every trial is a typed record that publishes only what its basis permits | delivery-economics-are-measured-over-declared-windows Delivery economics are measured on corpus runs and on real work over a window registered before it opens, and published with their denominators and losses | 1: I3352 → I3383 |
| the-proof-tests-are-registered-before-they-run The six proof tests are registered as validated protocols, each frozen by its own digest, and every trial is a typed record that publishes only what its basis permits | recovery-from-injected-faults-is-proven A killed worker, a cut connection and a conflict are injected on purpose, and recovery is proven without lost work or duplicate completion | 2: I3352 → I3391 I3353 → I3391 |
| the-proof-tests-are-registered-before-they-run The six proof tests are registered as validated protocols, each frozen by its own digest, and every trial is a typed record that publishes only what its basis permits | the-corpus-is-registered-before-it-runs A corpus of 20 tasks, stratified by a difficulty proxy measured and registered before the new tasks are written, is registered with its analysis, run order, stopping rule and a pinned release | 2: I3351 → I3367 I3352 → I3370 |
| the-proof-tests-are-registered-before-they-run The six proof tests are registered as validated protocols, each frozen by its own digest, and every trial is a typed record that publishes only what its basis permits | the-mesh-is-measured-against-one-agent Coordinated workers on the mesh are compared with a single agent on the same batch tasks, at quality held by a registered non-inferiority guard | 1: I3352 → I3388 |
| the-proof-tests-are-registered-before-they-run The six proof tests are registered as validated protocols, each frozen by its own digest, and every trial is a typed record that publishes only what its basis permits | the-proof-scorecard-is-published One derived scorecard publishes every proof test's computed status, losses included, on every surface, and no value claim stands without a passed test | 3: I3351 → I3354 I3352 → I3358 I3353 → I3354 |
| the-runtime-supervises-itself The runtime says what is progressing, waiting or stalled, and repairs only what is safe | landing-survives-any-session Landing green work survives any session, any restart and any machine | 1: I3182 → I3912 |
| the-runtime-supervises-itself The runtime says what is progressing, waiting or stalled, and repairs only what is safe | prismatic-lumen Prismatic Lumen: the owner oversees and moves the whole mesh from one page, which owns nothing | 1: I3182 → I3138 |
| the-runtime-supervises-itself The runtime says what is progressing, waiting or stalled, and repairs only what is safe | the-mesh-is-operated-from-one-page The owner operates the mesh from one page: every runtime readable, one action authorised end to end | 1: I3182 → I3191 |
| the-session-domain-is-written-once The session lifecycle is written by the typed domain, and the shell only parses and renders | entry-proves-what-it-handed An episode records the governance and briefing it was handed, and a change is told | 1: I3142 → I3148 |
| truthful-plan-surfaces The plan's surfaces are inferred from delivery, and nothing they show is typed beside the repository | a-claim-stands-only-on-its-evidence A claim that something is enforced, tested or delivered stands only on recorded evidence that would have caught it being false | 1: I2740 → I3119 |
| truthful-plan-surfaces The plan's surfaces are inferred from delivery, and nothing they show is typed beside the repository | a-visitor-understands-and-checks-majordomus A first-time visitor of majordomus.dev understands what Majordomus does and can check every claim it makes | 1: I2745 → I3224 |
| truthful-plan-surfaces The plan's surfaces are inferred from delivery, and nothing they show is typed beside the repository | command-and-deployment-criteria-are-judged A command or deployment criterion is met by a recorded run of exactly that command, or an observation of exactly that deployment, current at the presented revision | 1: I2740 → I3257 |
| truthful-plan-surfaces The plan's surfaces are inferred from delivery, and nothing they show is typed beside the repository | delivery-flow-is-measured The way from a pull request to a verified publication is measured per phase, and its bounds are read where work starts | 1: I2740 → I3203 |
| truthful-plan-surfaces The plan's surfaces are inferred from delivery, and nothing they show is typed beside the repository | merging-and-referencing-are-enforced Merging and referencing are enforced: every pull request carries its metadata, every reference resolves, every number is taken once | 3: I3449 → I3977 I3754 → I3971 I3757 → I3972 |
| truthful-plan-surfaces The plan's surfaces are inferred from delivery, and nothing they show is typed beside the repository | one-kit-on-every-surface Code, charts, controls, embeds and shell blocks are one kit on every surface, declared in the design book and held there by a gate | 2: I2747 → I3790 I2747 → I3791 |
| truthful-plan-surfaces The plan's surfaces are inferred from delivery, and nothing they show is typed beside the repository | the-plan-says-what-a-fix-unblocks The plan says what finishing an issue unblocks, and why each blocked issue is blocked | 1: I2740 → I3270 |
| web-sign-in The deployed surfaces know who is asking, and the identity providers are data | prismatic-lumen Prismatic Lumen: the owner oversees and moves the whole mesh from one page, which owns nothing | 1: I1205 → I3139 |
| web-sign-in The deployed surfaces know who is asking, and the identity providers are data | the-mesh-is-operated-from-one-page The owner operates the mesh from one page: every runtime readable, one action authorised end to end | 1: I1205 → I3192 |
| what-a-workspace-decides-reaches-the-repository-with-its-source What the owner decides in a vendor workspace reaches the repository, and every promoted record opens the message it came from | a-promoted-record-opens-its-message Every record promoted from a workspace cites the message it came from, and the UI opens it for the operator who holds it | 2: I3278 → I3284 I3279 → I3284 |
| what-a-workspace-decides-reaches-the-repository-with-its-source What the owner decides in a vendor workspace reaches the repository, and every promoted record opens the message it came from | intent-intake A received brief is a campaign revision whose requirements are reconciled into the one plan | 1: I3278 → I3273 |
The execution waves, every issue in the layer its dependencies put it in, are on the plan. Read the same thing at a terminal with majordomus plan graph and majordomus plan waves.