Module builtin

Expand description

The executable capabilities this executable ships: one Rust module per capability module, each owning its typed inputs and outputs, its handlers, its benchmark cases and its module() composition; modules composes the application from them with compose_modules!, and that list is the only root composition there is. Adding a capability to an existing module touches that module’s file alone; every projection, benchmark target and generated document follows from the descriptor.

use majordomus_cli::capability::builtin;

// the application is its modules, and each module stamps its namespace on what it holds
let modules = builtin::modules();
assert!(modules.iter().any(|m| m.id.as_str() == "repository"));
for module in &modules {
    for e in &module.capabilities {
        assert_eq!(
            e.capability.id.namespace(),
            module.id.as_str(),
            "{} is composed into the wrong module",
            e.capability.id
        );
    }
}

// and `all` is the same set flattened, for a registry built without module metadata
let flattened = builtin::all().len();
assert_eq!(
    flattened,
    modules.iter().map(|m| m.capabilities.len()).sum::<usize>()
);

Re-exports§

pub use artifacts::ArtifactReport;
pub use artifacts::ArtifactState;
pub use artifacts::ArtifactTallies;
pub use artifacts::ArtifactVerdict;
pub use artifacts::ArtifactView;
pub use artifacts::ArtifactsInput;
pub use artifacts::ARTIFACTS_URI;
pub use continuity::ActiveTask;
pub use continuity::Continuity;
pub use continuity::Divergence;
pub use continuity::OpenSession;
pub use continuity::Record;
pub use continuity::CONTINUITY_URI;
pub use devtask::DevMilestoneInput;
pub use devtask::DevTaskInput;
pub use environment::EnvironmentInput;
pub use environment::EnvironmentProvenance;
pub use environment::ExplainInput;
pub use environment::ENVIRONMENT_URI;
pub use episodes::AttachInput;
pub use episodes::DetachInput;
pub use episodes::EpisodeList;
pub use gates::CompletionInput;
pub use gates::GateModelEntry;
pub use gates::GateModelReport;
pub use gates::COMPLETION_URI;
pub use gates::GATES_URI;
pub use health::Health;
pub use health::HealthCheck;
pub use health::HealthStatus;
pub use health::HEALTH_URI;
pub use knowledge_base::AgeSource;
pub use knowledge_base::KnowledgeCandidate;
pub use knowledge_base::KnowledgeCandidates;
pub use knowledge_base::KnowledgeRecord;
pub use knowledge_base::KnowledgeRecordInput;
pub use knowledge_base::KnowledgeStatus;
pub use knowledge_base::LedgerMark;
pub use knowledge_base::ReferenceResolution;
pub use knowledge_base::ResolvedReference;
pub use knowledge_base::ResolvedRelation;
pub use knowledge_base::KNOWLEDGE_CANDIDATES_URI;
pub use knowledge_base::KNOWLEDGE_STATUS_URI;
pub use lifecycle::Balance;
pub use lifecycle::ClosedSession;
pub use lifecycle::ClosedSessions;
pub use lifecycle::Episode;
pub use lifecycle::EpisodeStanding;
pub use lifecycle::Episodes;
pub use lifecycle::Orphan;
pub use lifecycle::Pointer;
pub use lifecycle::PointerLayout;
pub use lifecycle::ProviderLifecycle;
pub use lifecycle::ProviderLifecycles;
pub use lifecycle::Recovery;
pub use lifecycle::RuntimeView;
pub use lifecycle::Stranded;
pub use lifecycle::EPISODES_URI;
pub use lifecycle::RECOVERY_URI;
pub use objects::resolve;
pub use objects::AnswerView;
pub use objects::Comparison;
pub use objects::DriftedObject;
pub use objects::GetInput;
pub use objects::ListInput;
pub use objects::ObjectList;
pub use objects::ObjectStanding;
pub use objects::Resolved;
pub use objects::ResourceView;
pub use objects::SearchHit;
pub use objects::SearchInput;
pub use objects::SearchResult;
pub use objects::VerifyInput;
pub use objects::VerifyReport;
pub use objects::SEARCH_DEFAULT_LIMIT;
pub use objects::SEARCH_MAX_LIMIT;
pub use obligations::Closure;
pub use obligations::Evidence;
pub use obligations::Obligation;
pub use obligations::ObligationClosure;
pub use obligations::ObligationState;
pub use obligations::Vocabulary;
pub use obligations::CLOSURE_URI;
pub use obligations::OBLIGATIONS_URI;
pub use plan::PlanIssueFilter;
pub use plan::PlanMilestoneFilter;
pub use plan::PlanRecordInput;
pub use plan::PLAN_URI;
pub use quality::QualityAnswer;
pub use quality::QualityInput;
pub use quality::QUALITY_URI;
pub use repository::RepositoryReport;
pub use repository::REPOSITORY_URI;
pub use server::Checkouts;
pub use server::Desired;
pub use server::LeaseView;
pub use server::ServerStanding;
pub use server::ServerStatus;
pub use server::ServerStatusInput;
pub use server::ServerView;
pub use server::SERVER_URI;
pub use session_domain::IdentityReport;
pub use session_domain::MachineReport;
pub use session_domain::IDENTITY_URI;
pub use session_domain::MACHINE_URI;
pub use trace::TraceCommitInput;
pub use trace::TraceIssueInput;
pub use trace::TraceReportInput;
pub use trace::TRACEABILITY_URI;
pub use commands::CommandGraphReport;
pub use commands::CommandIndex;
pub use commands::CommandSummary;
pub use web::SurfaceReport;
pub use web::SURFACES_URI;

Modules§

artifacts: The artifacts module: what majordomus generate writes, read back.
commands: The commands module: the command graph, projected.
commit: The commit module: what this repository holds a commit to, what the working tree would commit, and which scope the history gives a set of paths.
continuity: The continuity module: what this checkout’s lifecycle is holding right now — the open episode, the record the next worker would resume from, the newest progress note, what is blocking acceptance, and whether the provider’s events actually reach the tool.
dashboard: The dashboard module: the Overview of the Dashboard Suite, as a projection of the capabilities that already answer each of its facts (ADR 0088).
devtask: The devtask module: an issue and a milestone read as executable development scopes.
economics: The economics module: token economics, measured, projected. Every capability here is a read of crate::economics — the one calculator — so the command line, HTTP, the OpenAPI document, MCP and the Cockpit show the same metric, value, class and sample count, because none of them computes one.
entity: The entity module: every object of the layer as an addressable, cross-linked node.
environment: The environment module: what this checkout is, as a capability.
episodes: The episodes module: the execution episode of a client that has no provider hooks, opened and closed by the MCP connection itself (ADR 0103).
evidence: The evidence module: from a claim to the run that proves it, and back.
gates: The gates module: completion as a state the repository decides.
health: The health module: whether what this process serves is healthy, answered by the engines that already decide it rather than by checks written a second time here.
intent_realization: The intent_realization module: which work realises which intent, and why an intent stands where it stands.
intents: The intents module:
knowledge_base: The knowledge_base module: what the knowledge deriver left for review and whether it is still writing — the candidate records awaiting promotion, one record by id with every reference it names resolved, and the derivation status of this checkout judged against session.freshness.
lifecycle: The lifecycle module: the episode lifecycle as something a person can watch.
mcp: The mcp module: the MCP projection of this registry, described as one typed value.
objects: The objects module: the declarative objects of the repository’s layer, listed, read and searched, and the one resolution of a majordomus:// URI that objects.get and the MCP resource read share.
obligations: The obligations module: what a task owes before it may be called completed, and whether the evidence that discharged each one still describes this tree.
observatory: The observatory module: what the Observatory (/ui/observatory) may carry to a place where no mesh runs, and the recording the repository carries there (ADR 0158, I2678).
pack: The pack module: the tracked tree as token-bounded text shards for a reader whose file search indexes text, with nothing in it that is not source.
plan: The plan module: the milestone and issue model’s derivations, projected.
quality: The quality module: what this executable’s own public surface is held to, answered through every transport the surface is measured against.
reasoning: The reasoning module: provider-independent reasoning, projected (ADR 0098).
recover: The recover module: the stray files of the record stores, classified by evidence and swept exactly once.
release: The release module: the changelog, the version, and the compatibility analysis behind it — all three derived.
repository: The repository module: what the repository and this process’s index are.
rules: The rules module: from a rule to what proves it, and back.
served: The served module: whether a deployment serves the commit it was meant to, observed from outside and recorded as evidence a later reader can re-judge.
server: The server module: the shared server of this checkout, and of every other checkout of the same git repository — what each lease says, whether the server it names answers, and whether what answers is the code on disk.
session_domain: The session_domain module: the typed session domain, projected.
skills: The skills module: every skill as a proven capability, derived.
trace: The trace module: the work graph above the branch, both ways.
web: The web module: the surfaces this repository exposes over HTTP, answered from the resolution the process is already serving from.

Structs§

AnnounceInput: The input of peers.announce: what the calling peer is working on.
BuildReport: What this executable is: enough to tell two builds apart without asking a repository.
CancelReport: The answer of executions.cancel.
CapabilitiesInput: The input of capabilities.list: optional filters by kind and by projection.
CapabilityList: The answer of capabilities.list: the matching capabilities, summarised, and the registry counted.
CapabilitySummary: One capability as a listing shows it: everything the descriptor says except its two schemas, which capabilities.describe answers for one capability. A listing of a repository’s registry runs to hundreds of entries; their schemas would be megabytes of the same object view repeated.
ClassifyInput: Which path to judge.
ContractView: A directory contract as the document declares it, before anything is inherited.
DeploymentCheck: Whether the deployments this repository declares would work, decided locally: against the capability registry this process built and the workspace it sits in.
DeploymentList: Every deployment the layer declares.
DeploymentView: One deployment as this process reads it, with where it came from.
DescribeInput: The input of capabilities.describe: which capability.
DirectoriesInput: Which directories to answer for, and how much of the hierarchy to resolve.
DirectoryNode: One directory of the layer, what it owes, and what applies to it.
DirectoryReport: The layer’s directories, what each owes, and the contracts that apply.
DirectoryTallies: How many directories are in each state.
DistributionReport: The distribution, as a person or a script asks for it.
EffectiveEntry: One contract in a directory’s effective chain, in the order it is applied.
Empty: No input.
EventHistory: The answer of executions.events: one page of an execution’s retained history.
ExecutionLinks: Where to read more about one execution, in this server’s own terms.
ExecutionList: The answer of executions.list.
ExecutionView: One execution with the links to everything else about it.
GetDeploymentInput: The input of deploy.get: which deployment.
GraphInput: Which graph to derive.
GraphList: Every graph this executable derives.
InstallCheck: One check: what was asked, what was seen, and — when it does not hold — why, and the command that changes it. The cause and next fields exist so that a report is actionable without a second document; a check that fails without naming its remedy is a diagnostic nobody can act on.
InstallabilityReport: Whether the advertised one-line installation works right now, and if not, what is missing.
MeshIdentityReport: The answer of mesh.identity: this machine’s node identity, public half only.
ModelsFilter: The input of models.list: optional narrowing. Everything empty lists everything.
ModelsReport: The answer of models.list.
NodeList: The answer of mesh.nodes: every node this process has observed.
ObjectSummary: One object, summarised for a listing.
ObjectView: One declarative object of the repository’s layer, as a client reads it.
PeerList: The answer of peers.list: every worker of this repository, from every checkout’s board.
ProtocolReport: The answer of executions.protocol: the live channel’s whole contract, derived from the types that implement it.
RegisterInput: The input of mesh.register: one signed envelope, exactly as the wire carries it.
ReleaseArtifactInput: Which artifact a target and a tag name.
ReleaseArtifactView: The artifact a target and a tag name, and the directory it unpacks into.
ReleaseView: One release, as every projection shows it.
ReleasesReport: The releases this repository recorded, and the one an unpinned installation resolves to.
RouteInput: The input of models.route: what the task needs, in transport-friendly scalars (require is comma-separated so a GET query string can carry it).
ScopeReport: The scope: the declaration as read, its origin, and every tracked file against it.
TargetView: One target, as every projection shows it: what it is, what it is called in prose, and the artifact name the naming function derives for a tag yet to be chosen.
VendorView: One vendor with whether its credential is configured — presence of the named environment variable and nothing more. No value is read.
WorktreeInspectInput: A branch to inspect.
WorktreeStatusInput: Where to answer from.

Enums§

CheckState: The state of one check in the installability report. Three states and no more: a check either holds, does not, or could not be made from what this process can see.
DirectoryState: Whether a directory of the layer carries the contract it owes.

Constants§

DEPLOYMENTS_URI: The URI under which the deployments are read as one MCP resource.
DEVCONTEXT_POLICY_URI: The URI under which the compiler’s own rules are read as an MCP resource.
DIRECTORIES_URI: The URI under which directories.list is read as an MCP resource.
EXECUTIONS_URI: The MCP resource the list of executions is read at.
EXECUTION_PROTOCOL_URI: The MCP resource the live channel’s own contract is read at.
GRAPHS_URI: The URI under which graph.list is read as an MCP resource.
MESH_URI: The MCP resource mesh.status is projected as.
MODELS_URI: The MCP resource models.list is projected as.
SCOPE_URI: The URI under which repository.scope is read as an MCP resource.
WORKTREES_URI: The URI under which the topology is read as an MCP resource.

Functions§

all: Every executable of every module, flattened, for a registry built without module metadata (tests, benchmarks, doctests). The application builds from modules.
modules: The application: its modules, in one place. A new module is one line here; a new capability in an existing module is no line here.
normalise_path: A repository-relative path as the scope judges it: forward slashes, no ./, no trailing slash. Absolute paths and parent segments are refused.

Module builtin

Module builtin Copy item path

Re-exports§

Modules§

Structs§

Enums§

Constants§

Functions§

Module builtin