Original
Goal Decomposition
Turn a broad engineering goal into independently reviewable outcomes, explicit uncertainty, dependency edges, and a bounded first execution wave.
Agent orchestrationexperimental
← All bundles
15 CANONICAL SKILLS · V0.2.0
Agent Orchestration
Bounded goals, capability-aware routing, coordinated dispatch, recoverable attempts, and reviewed integration.
Inspect the selection
View manifest ↗Run from a full toolkit checkout. This prints a file manifest, not an installer.
node tools/bundle.mjs resolve agent-orchestrationOriginal
Agent Capability Discovery
Build an expiring capability inventory from actual worker and tool probes while separating advertised, observed, unavailable, and authorized operations.
Agent orchestrationexperimental
Original
Agent Task Routing
Match ready tasks to qualified workers using required capabilities, phase, effect policy, availability, and measured tradeoffs without hardcoding provider superiority.
Agent orchestrationexperimental
Original
Parallel Dispatch Planning
Select genuinely ready parallel work without colliding on shared files, mutable resources, unstable contracts, or review capacity.
Agent orchestrationexperimental
Original
Subagent Execution Contract
Give a delegated worker a bounded goal, accepted source identity, explicit tools/effects, resource limits, check requirements, and a result envelope.
Agent orchestrationexperimental
Original
Subagent Lifecycle
Track allocated, running, submitted, accepted, failed, and indeterminate work with explicit ownership and recoverable interruption.
Agent orchestrationexperimental
Original
Orchestration Resume
Resume interrupted coordination from checked run records, reusing only current accepted artifacts and withholding ambiguous or stale work.
Agent orchestrationexperimental
Original
Agent Failure Recovery
Classify agent failures, preserve evidence, and choose bounded retries or reconciliation without repeating costly or unsafe effects.
Agent orchestrationexperimental
Original
Agent Result Integration
Accept worker results against independent contracts and integrate combined changes without treating individual green branches as release evidence.
Agent orchestrationexperimental
Original
Orchestration Observability
Capture task/attempt traces that explain delegation, queue time, context identity, resource pressure, and whether submitted work was actually integrated.
Agent orchestrationexperimental
Original
Resource-aware Agent Scheduling
Budget model workers separately from CPU, memory, database, browser, and build resources while reserving capacity for acceptance and integration.
Agent orchestrationexperimental
Original
Work Packet Planner
Turn an engineering request into a bounded work packet with owned paths, dependencies, acceptance criteria, verification commands, and a handoff contract. Use before delegating non-trivial repository work.
Planningexperimental
Original
Worktree Handoff
Prepare a commit-specific engineering handoff from an existing Git worktree, separating actual changes and executed checks from unresolved work, resource ownership, and integration risks.
Parallel engineeringexperimental
Original
Bounded Context Assembly
Assemble task-sized context from current canonical source, accepted contracts, graphs, and semantic tools while tracking freshness and missing information.
Parallel engineeringexperimental
Original
Specialist Role Composition
Compose a small task-relevant specialist preset from canonical skills, with explicit outputs and non-goals while leaving permissions and execution to the host.
Role compositionexperimental
Tailor context, not a second copy.
Pick the skill needed for the current task. A private adapter supplies project-specific roots, command IDs, accepted contracts, and policy. It should pin and verify the public revision before exposing guidance to an internal agent.
Skill effectiveness remains experimental. Package validation and content hashes do not establish host compatibility, artifact authenticity, or permission to execute.
Read the private-adapter boundary ↗