Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/emaraschio/cursor-commandsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/emaraschio/cursor-commands/automation-roi-audit)<a href="https://agentmods.dev/commands/emaraschio/cursor-commands/automation-roi-audit"><img src="https://agentmods.dev/badge/commands/emaraschio/cursor-commands/automation-roi-audit/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/emaraschio/cursor-commands/automation-roi-audit"><img src="https://agentmods.dev/badge/commands/emaraschio/cursor-commands/automation-roi-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00040 | $0.00970 |
| Opus 5 | $0.00020 | $0.00485 |
| Sonnet 5 | $0.00008 | $0.00194 |
| Haiku 4.5 | $0.00004 | $0.00097 |
Grade A, and why
automation-roi-audit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Overview
Act as an AI operations consultant: intake business + function + workflow list, interview how each process runs, label steps, prioritize by money/margin/output, warn on traps, recommend one narrow one-week test. Audit only; does not implement the automation in the same turn. Full workflow: .cursor/skill-contracts/automation-roi-audit/SKILL.md (user install: ~/.cursor/skill-contracts/automation-roi-audit/SKILL.md).
Defaults
| Setting | Default |
|---|---|
| Profile | generic (sales / marketing / ops / customer support / content / finance) |
| Software-team profile | Optional; eng / product / platform / support / content / finance-adjacent |
| Interview | Required until each workflow has enough detail to label steps |
| Deliverable | Chat report; optional save under docs/automation-roi-audits/ after confirm |
| Execution | Recommend only; no implementation in the same turn |
| Examples | Generic names only (business-a, workflow-1) |
Steps
- Read
.cursor/skill-contracts/automation-roi-audit/SKILL.mdfor the full agent contract; if that file is missing, read~/.cursor/skill-contracts/automation-roi-audit/SKILL.md. - Execute phases in order (Intake → Interview gate → Deconstruct and label → ROI / warnings / one-week test → Deliver); do not skip the interview gate or implement the recommended automation.
- Report the audit in chat; offer optional docs save only when the host has
docs/and the user confirms.
Anti-patterns
- Interview before analyzing. Trigger: user pastes a thin workflow list (names or one-liners only). Wrong: labeling steps and picking ROI immediately. Correct: interview until each workflow has enough process detail, then analyze. Reason: labels and ROI from thin lists invent process and ship bad bets.
- Reject AI theater as the primary bet. Trigger: a suggested automation that does not improve money, margin, or output. Wrong: ranking it as the highest-ROI opportunity. Correct: call it AI theater and pick a bet tied to money, margin, or output. Reason: theater burns time without changing the business.
- Warn on rebuild-cheap-SaaS and distraction traps. Trigger: the idea is a meeting-summary toy, a distraction, or rebuilding something like Calendly for pocket change. Wrong: recommending it as the one-week test. Correct: warn explicitly and steer to a narrower proprietary or high-leverage test. Reason: those traps look like progress and displace real ROI work.
- Prefer proprietary edge. Trigger: ranking automation opportunities. Wrong: generic wrapper automations ranked over process-specific leverage. Correct: tie the bet to proprietary data or domain workflow when available. Reason: proprietary edge is harder to copy and more likely to move money, margin, or output.
- Write docs only after confirmation. Trigger: host has a
docs/directory. Wrong: writingdocs/automation-roi-audits/...unprompted. Correct: offer the path and write only after the user confirms. Reason: unsolicited docs clutter the host workspace. - Do not implement the automation in the same turn. Trigger: finishing the audit with a clear one-week test. Wrong: scaffolding code, PRs, or a follow-on skill without a new ask. Correct: deliver the report (and optional confirmed docs save) and stop. Reason: this skill audits; building skips review of the recommendation.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 51 lines · 40 tokens per session scan A dc39e929ff8f
automation-roi-audit is a command published in the GitHub repository emaraschio/cursor-commands (9 stars, last pushed 1mo ago), licensed MIT. It adds 40 tokens to every session and 970 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
grok_execute
Explicitly turn the task-local Grok executor mode on or off.
grok_init
Discover, verify, and persist the local HTTP proxy used by Grok Build Supervisor.
handle-pr-comments
Fetch all PR comments, classify by severity, and present for user approval before acting.
self-review
Review only changed files in this branch Your primary goal is to provide valuable, trustworthy feedback while avoiding false positives and low-impact commentary.
create-pr
Commit changes and create a PR with proper formatting.
back-to-master-delete-branch
Cleanup current branch and switch to default branch with latest changes.