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.
npx skills add gguloadoong/idea-factory --skill start-companygit clone --depth 1 https://github.com/gguloadoong/idea-factoryWrote 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/skills/gguloadoong/idea-factory/start-company)<a href="https://agentmods.dev/skills/gguloadoong/idea-factory/start-company"><img src="https://agentmods.dev/badge/skills/gguloadoong/idea-factory/start-company/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/skills/gguloadoong/idea-factory/start-company"><img src="https://agentmods.dev/badge/skills/gguloadoong/idea-factory/start-company.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 | $0.00032 | $0.00794 |
| Opus 5 | $0.00016 | $0.00397 |
| Sonnet 5 | $0.00006 | $0.00159 |
| Haiku 4.5 | $0.00003 | $0.00079 |
Grade A, and why
start-company 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 11d 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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
idea-factory
You received a business idea: $ARGUMENTS
Build a virtual startup and deliver a working MVP.
Principles
- MVP-First: Prototype with mock data first. Validate value before real APIs.
- CEO = User: Never ask technical questions. Offer choices in plain language.
- Essence-Driven: Every feature must serve the core "Why" of the service.
- Zero-Prompting Goal: CEO gives the idea once. The team handles the rest.
- Heal, Don't Repeat: If something fails twice, find the root cause.
- Model Strategy: Sonnet default. Opus for architecture/RCA/deep analysis.
Execution Flow Overview
Seven phases run in sequence. Full per-phase detail is in phases.md.
- ANALYZE — Two parallel agents (analyst + architect, both opus) dissect the idea
- SCAFFOLD — Create project from
~/.claude/templates/company/(26 items) - KICKOFF — Ask CEO 3–5 plain-language questions; fill CLAUDE.md + PRD + essence.md
- BUILD MVP — Generate
.omc/prd.jsonstories; start/oh-my-claudecode:ralph - VALIDATE — 4 isolated reviewers (architect / critic / code-reviewer / qa-tester) per gate
- HARDEN — Real APIs, tests, security audit, protected-files, quality-baseline
- SHIP — GitHub + 5-stage PR pipeline + 6-gate deploy consensus + retro
For per-phase step-by-step instructions, see phases.md.
For agent role definitions and assignment rules, see agents.md.
Critical Invariants (apply across all phases)
CLAUDE.md Size Rule
- Generated CLAUDE.md MUST stay under 80 lines (leaves room for growth)
- NEVER replace CLAUDE.md with a single
@AGENTS.mdpointer - Use progressive disclosure: details in separate files, pointers in CLAUDE.md
- Reference: https://www.humanlayer.dev/blog/writing-a-good-claude-md
Fix-Loop Circuit Breaker
- If any story fails the same way 3 times → STOP and escalate to CEO
- After 3 failures: write diagnosis to
.project/decisions.mdand flag to CEO - This overrides "Heal, Don't Repeat" — healing has a budget
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 11d ago First seen · 76 lines · 32 tokens per session scan A 80fbb8804d3e
start-company is a skill published in the GitHub repository gguloadoong/idea-factory (2 stars, last pushed 4mo ago), licensed MIT. It adds 32 tokens to every session and 794 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 skills, from other repositories
ouroboros-run
Execute a Seed specification through the workflow engine.
auto
Automatically converge from goal to A-grade Seed and execute it.
evolve
Start or monitor an evolutionary development loop.
publish
Publish Seed specification as GitHub Issues for team-based project management.
tutorial
Interactive tutorial teaching Ouroboros hands-on.
ouroboros-status
Check session status and measure goal drift.