ROI — AI skill for Claude Code
Stage 08. Invoke the engagement-manager in roi mode to populate the engagement ROI model with actuals against the stage-04 baseline, and draft the executive readout.
How to install ROI
Installs to ~/.claude/commands/jpsusarreyhgs-fde-harness-roi.md
mkdir -p ~/.claude/commands && curl -fsSL https://raw.githubusercontent.com/jpsusarreyHGS/fde-harness/HEAD/.claude/commands/roi.md -o ~/.claude/commands/jpsusarreyhgs-fde-harness-roi.md Restart Claude Code, or start a new session, for it to be picked up.
What ROI does
description: Stage 08. Invoke the engagement-manager in roi mode to populate the engagement ROI model with actuals against the stage-04 baseline, and draft the executive readout. Nine inputs, four outputs, plus counter-metrics and an agreed attribution basis. allowed-tools: Read Write Glob Grep Bash Agent
**Before dispatching, run the pre-flight.** If it exits non-zero, show its message and stop — do not dispatch. An agent isolated in a fresh worktree cannot see an engagement folder, an
Alternatives in AI
- PRD: Agent Builder — Create Your Own Sub-Agent — Your ecosystem, your rules 1.9k ★
- Geminimcp — Gemini-MCP is an MCP server that encapsulates Google's Gemini CLI tool into a standard MCP protocol interface 294 ★
- Bounty Concierge — Start Here -- RustChain Bounty Concierge for humans and AI agents 110 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Sow
把当前会话围绕给定意图浓缩并归档到用户 vault, AI 判层 (Inbox / Inputs / Outputs) + 用户 NL 确认
Allocate
Stage 04. Invoke the solution-architect in allocate mode to place the intelligence — assign every operating-ma
Ontology
Stage 03. Invoke the ontology-engineer to promote discovery output into the client domain model. Assembles the
Answer Contract
Four rules that stop your AI agent from padding, rambling, and shipping one mediocre answer: Implicit-Need Fil
Tauri Plugin MCP
Allows AI agents (e.g., Cursor, Claude Code) to debug within Tauri apps via screenshot capture, window managem
Agent Runtime
Provider-agnostic LLM agent loop for Node: bounded tool calling with validated inputs, token-budget memory, tr
Related Agents
Huella Reporter
Assembles the bilingual (Spanish) YOUR_ORG "Informe de Huella Digital" for a huella_digital engagement. Reads
Executive Review Agent
Strategic business review agent with executive perspective who evaluates projects from ROI, market impact, res
Draft Commander
Lead AI agent and Head Coach of a 5-agent Fantasy Football War Room — synthesizes sub-agent inputs and makes t