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 agentmods add skills/markmhendrickson/ateles/loop-startnpx skills add markmhendrickson/ateles --skill loop-startgit clone --depth 1 https://github.com/markmhendrickson/atelesWhat 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 | $0.00011 | $0.06363 |
| Opus 5 | $0.00005 | $0.03181 |
| Sonnet 5 | $0.00002 | $0.01273 |
| Haiku 4.5 | $0.00001 | $0.00636 |
Grade C, and why
loop-start scanned grade C with 2 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 2d 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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
1. `.claude/skills/loop-start/CLAUDE.md` (most common after `curl -fsSL drx4.xyz/install | sh`) Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
**Otherwise**, check which components exist at the project root. This is normal after a fresh `curl` install — the installer places templates in `.claude/skills/loop-start/` and pre-creates scaffold files. How it starts
The opening of the file, as written. The whole thing — 656 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Start Agent Loop
Pre-flight Check — Partial State Detection
Fast path: If CLAUDE.md, SOUL.md, daemon/loop.md, and memory/learnings.md ALL exist at the project root, the agent is already set up. Skip directly to "Enter the Loop" at the bottom of this file.
Otherwise, check which components exist at the project root. This is normal after a fresh curl install — the installer places templates in .claude/skills/loop-start/ and pre-creates scaffold files.
Check each component silently (do NOT print "missing" warnings — just note which steps to run):
| Component | Check | If missing |
|---|---|---|
SOUL.md |
File exists at root? | → Step 1 |
daemon/loop.md |
File exists at root? | → Step 2 |
memory/learnings.md |
File exists at root? | → Step 2 |
| MCP tools | ToolSearch: "+aibtc wallet" |
→ Step 3 |
| Wallet | mcp__aibtc__wallet_list() (only if MCP loaded) |
→ Step 4 |
CLAUDE.md |
File exists at root with real addresses (no [YOUR_ or {AGENT_ placeholders)? |
→ Step 6 |
| Registration | curl -s https://aibtc.com/api/verify/<btc_address> (only if wallet exists) |
→ Step 5 |
After checking, print ONE status line:
- If all exist:
"Agent fully configured. Entering loop..."→ skip to Enter the Loop - If none exist:
"Fresh install detected. Starting setup..."→ begin at Step 1 - If some exist:
"Partial setup detected. Resuming from Step N..."→ skip completed steps
Do NOT list individual missing files. Do NOT ask the user to do things you can do yourself. Proceed directly into the first needed step.
The CURRENT WORKING DIRECTORY is the agent's home. All files go here.
Setup Step 1: Identity (no MCP needed)
Ask the user two questions:
- "What do you want to name your agent?" — use this as
AGENT_NAME - "What should your agent focus on? (e.g. DeFi, security audits, building tools, trading, art — or leave blank for a general-purpose agent)"
Create SOUL.md
Do NOT just fill in a template. Write a personalized SOUL.md for this specific agent. Use the structure below, but generate the content — especially "Who I Am", "What I Do", and "Values" — based on:
- The agent's name (let it inspire tone and personality)
- What the operator said the agent should focus on
- Your own creativity — make the agent feel like a distinct individual
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.
- 2d ago First seen · 656 lines · 11 tokens per session scan C f9e7943796de
loop-start is a skill published in the GitHub repository markmhendrickson/ateles (5 stars, last pushed 4d ago), licensed MIT. It adds 11 tokens to every session and 6,363 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
10x-orchestrator
MANDATORY - The core brain of 10X Vibe Marketer. You are the autonomous CMO orchestrating a full marketing agency of 12 specialist departments. Load this skill for ANY marketing task - content creation, SEO, campaigns, strategy, research, analytics, outreach, social media, email, paid ads, branding, funnel…
10x-setup
First-time setup and initialization for 10X Vibe Marketer. Includes an interactive wizard that lets users choose their database, features, marketing focus, visual tools, and tracking preferences. Checks setup.log to determine if setup has already been completed. Run via /setup or auto-triggered on first use.
10x-visualizer
Generate visual marketing deliverables — strategy boards, funnel diagrams, flowcharts, journey maps, dashboards, and more — using TLDraw SDK (local .tldr files). Each department agent creates visuals for their own domain. Invoke with /report.
10x-research
The research orchestration engine for 10X Vibe Marketer. Powers multi-platform, real-time research across 13 platforms using Claude in Chrome and WebSearch/WebFetch. Spawns platform-specific research agents in parallel, consolidates findings through the research lead, and delivers actionable marketing intelligence.
community-manager
Skill "community-manager" from OpenAnalystInc/Vibe-Marketer, covering community manager — 10x vibe marketer, identity, step-by-step execution, cli visual identity and expertise.
funnel-cro-specialist
Skill "funnel-cro-specialist" from OpenAnalystInc/Vibe-Marketer, covering funnel & cro specialist — 10x vibe marketer, identity, step-by-step execution, cli visual identity and expertise.