dynos-work

Run checkpoint audit, repair any findings, then reach DONE — all in one shot. Use after /dynos-work:execute.

Execute the approved plan. Orchestrates execution graph segments through specialized executor agents, including dependency management and error recovery.

Internal: Data Executor. Implements ETL, analytics, backfills, data quality, reconciliation, and data pipeline changes. Write evidence on completion.

Internal: Infrastructure Executor. Implements deployment, CI/CD, container, IaC, and environment configuration changes. Write evidence on completion.

Internal: Observability Executor. Implements logs, metrics, traces, alerts, dashboards, and reliability instrumentation. Write evidence on completion.

Internal: Release Executor. Implements version/changelog updates, rollout flags, rollback plans, release notes, and migration sequencing. Write evidence on completion.

Internal: Security Executor. Implements auth, authorization, validation, crypto, secret-handling, and vulnerability remediations. Write evidence on completion.

Internal dynos-work executor skill group. Routes execution graph segments to the specialized executor skills.

Internal dynos-work skill. Agent calibration — generates project-specific agents, benchmarks them, promotes/demotes based on performance, and manages auditor mode transitions.

Internal dynos-work skill. Validate contract.json chaining across the skill pipeline. Checks that output_schema of each pipeline stage covers the required input_schema fields of the next stage.

Deep bug investigation. Pass a short description of the problem — error message, unexpected behavior, or failing test. Returns structured root cause analysis with evidence and fix recommendation.

Internal dynos-work skill. Re-run planning on an existing task. Use when you need to regenerate the plan after spec changes, or if the task was started externally. Runs PLANNING → PLAN_REVIEW → PLAN_AUDIT with deterministic artifact validation.

Internal dynos-work skill. Register the current project directory with the dynos-work global registry. Works from any project.

Inspect and drain the proactive-residual queue at .dynos/proactive-findings.json. Subcommands: list, run-next.

Resume an interrupted task from .dynos/ state. Use after session restart or context compression.

Start a new task. Give it a description and it handles discovery, spec, plan, execution, and audit.

Show current task state, lifecycle stage, audit results, and open gaps.

Internal dynos-work skill. Start the SPA dashboard server showing all registered projects.

Internal dynos-work skill. Manage the global cross-project sweeper daemon.

Internal dynos-work skill. Manually repair a specific finding. Use when you want to fix one issue without running the full audit loop. /dynos-work:audit handles repair automatically.

Internal: Backend Executor. Implements API routes, services, business logic, auth. Read spec and segment. Write evidence on completion.

Internal: DB Executor. Implements schema changes, migrations, ORM models, queries. Read spec and segment. Write evidence on completion.

Internal: Integration Executor. Wires components together, connects external APIs, handles plumbing. Write evidence on completion.

Internal: ML Executor. Implements ML models, training pipelines, inference code, data processing. Read spec and segment. Write evidence on completion.

Internal: Refactor Executor. Restructures code without changing behavior. No new features. Write evidence on completion.

Internal: Testing Executor. Writes unit, integration, and e2e tests. Read spec and segment. Write evidence on completion.

Internal: UI Executor. Implements UI components, pages, interactions, and styles. Read spec and segment. Write evidence on completion.

Internal dynos-work skill. Set up the current project: register with dynos-work and start the local daemon.

Internal dynos-work skill. List all registered dynos-work projects.

Internal dynos-work skill. Manage the project daemon: start, stop, status, logs, run-once, dashboard.

Internal dynos-work skill. Autonomous maintenance worker. Periodically scans the repo for technical debt, security anti-patterns, and architectural drift. If issues are found, it automatically creates a repair branch, implements the fix, verifies with audits/tests, and opens a Pull Request.

Internal dynos-work skill. Sequence memory manager. Stores compact task traces and retrieves similar prior tasks to inform discovery and design review.

Internal dynos-work skill. Aggregate task retrospectives into project memory. Scans task-retrospective.json files, writes project_rules.md, computes effectiveness scores, derives Model Policy and Skip Policy, manages baselines. Runs automatically at task completion.