Productivity & Workflow Agent Skills for Devin Desktop

Delegate a coding task to the Grok Build CLI as a background implementer, then review its diff and land it yourself.

Use when assigning backlog items to a GitHub milestone.

Surface Claude's assumptions about a phase approach before planning

List active GSD workspaces and their status

Automate Habitica tasks via Rube MCP (Composio).

Write a session handoff that lets a fresh agent (or you after compaction) resume cold with zero re-discovery.

The meta-skill above the per-engine adapters — pick the right coding-agent harness for a goal, drive fleets of headless instances through…

Use when asked to build a product roadmap, prioritize a backlog, decide what to build next, or sequence a list of feature ideas.

Audit and clean ~/.hermes safely by separating core state from caches, logs, and rebuildable artifacts.

Use when provisioning Hermes worker profiles.

Manage and clean up Hermes Desktop Project workspaces.

Build, modify, validate, and troubleshoot importable HiAgent workflow JSON from user-provided exports or verified node samples.

Run a multi-model adversarial mutual review with facilitator synthesis: 2+ panelists (different model families where possible — use…

Orchestrate multiple coding-agent sessions (Claude Code, Codex, Gemini) over the mcp_agent_mail MCP server, observe their progress on a…

Run OpenAI Codex CLI from the terminal for batch / one-shot `codex exec` invocations with output captured to an `-o` file.

Orchestrate free opencode workers from Claude Code to cut token costs.

Executes approved implementation plans safely with dual verification profiles and independent reviews.

Track long-horizon objectives across multiple sessions with milestone checkpoints, progress persistence, and drift detection

How the team works on Laravel projects — repository reconnaissance before making any changes, best practices during implementation, and…

HR business partnership across talent strategy, org development, and employee relations.

Design human-in-the-loop (HITL) approval gates for AI agent workflows.