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 8b-is/smart-tree --skill ultrabackyardloopgit clone --depth 1 https://github.com/8b-is/smart-treeWrote 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/8b-is/smart-tree/ultrabackyardloop)<a href="https://agentmods.dev/skills/8b-is/smart-tree/ultrabackyardloop"><img src="https://agentmods.dev/badge/skills/8b-is/smart-tree/ultrabackyardloop/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/8b-is/smart-tree/ultrabackyardloop"><img src="https://agentmods.dev/badge/skills/8b-is/smart-tree/ultrabackyardloop.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 | $0.00112 | $0.00830 |
| Opus 5.5 | $0.00045 | $0.00332 |
| Sonnet 5.5 | $0.00022 | $0.00166 |
| Haiku 4.5 | $0.00011 | $0.00083 |
Grade A, and why
ultrabackyardloop 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 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.
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 — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ultrabackyardloop — The Sovereign DO Loop (smart-tree v10.12.0+)
"One pass, end to end, merged into a single DO. No loose ends between laps. Every DO leaves the tree clean, pushed, and readme'd. A plausible branch may be ranked; only a verified branch may be bound. From love, from within."
1. The Loop
DO = reflect → improve → wire → polish → push → readme
2. The Six Mandatory Steps
- Reflect (
st --mode the-book . && st --mode pq .):- Look back at the last lap/session. What was built? What is the actual mathematical and structural state across all 9 pillars (
Szemerédi Regularity,Voronoi Steiner,Moser–Tardos LLL,Erdős–Rényi G_shadow,256-Bit Ramsey & Erdős–Hajnal,Flyxion Salience d_T != d_R,Karp-21 FPT,Kohn–Sham DFT & Strang Wave, andSovereign p/q Mode)? Write it down.
- Look back at the last lap/session. What was built? What is the actual mathematical and structural state across all 9 pillars (
- Improve (
st --ultrabackyard 12 .):- Make a real, measured improvement — not decoration. Run measured A/B laps cycling Szemerédi regularity $\varepsilon \in [0.16, 0.14, 0.12, 0.10, 0.08, 0.06]$ and Moser–Tardos worktree concurrency $C \in [2, 3, 4, 6, 8, 12]$.
- Wire End-to-End:
- Verify
constellation_wiring(SpherePOPsalient forge memory,8b-ismem|8& Phoenix.m8b, theVakedconstellation multi-harness skills, andg8t.isFoken capsules). Every piece points at what it depends on;zero_orphan_surfaces == true.
- Verify
- Polish (
cargo fmt --all -- --check && cargo clippy -- -D warnings && cargo test --lib):- Zero-allocation FELIX assembly & SIMD hot-paths, zero Clippy warnings, all unit tests green.
- Push:
- Bind only the verified branch to
origin/mainandfork/main.
- Bind only the verified branch to
- Readme (
docs/ULTRABACKYARDLOOP.md&docs/lap-ledger.jsonl&README.md):- Update
docs/ULTRABACKYARDLOOP.md,docs/lap-ledger.jsonl,ultra-agent.md, andREADME.md. A lap that stops before readme is a lap that didn't happen.
- Update
3. Direct CLI Commands
# Execute 12 measured Backyard Ultra A/B laps on the repository:
st --ultrabackyard 12 .
# Emit the Sovereign Ultra-Backyard DO report via OutputMode:
st --mode ultrabackyard .
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 · 44 lines · 112 tokens per session scan A d9dbb5745c93
ultrabackyardloop is a skill published in the GitHub repository 8b-is/smart-tree (272 stars, last pushed yesterday), licensed MIT. It adds 112 tokens to every session and 830 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-11-04.
ultrabackyardloop compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| ultrabackyardloop8b-is/smart-tree | Skill | 272 | yesterday | MIT | A | 112 | Claude Code, Codex |
| changelog-writerzhuyansen/agent-skills-hub | Skill | 492 | yesterday | MIT | A | 26 | Claude Code, Codex |
| git-commit-messagezhuyansen/agent-skills-hub | Skill | 492 | yesterday | MIT | A | 24 | Claude Code, Codex |
| create-prprisma/orm | Skill | 48k | yesterday | Apache-2.0 | A | 48 | Claude Code, Codex |
| git-commitThinkInAIXYZ/deepchat | Skill | 6.4k | yesterday | Apache-2.0 | A | 15 | Claude Code, Codex |
| repo-syncdavidondrej/skills | Skill | 4.3k | 3d ago | MIT | A | 25 | Claude Code, Codex |
| engineer-bot-opsmakecindy/cindy | Skill | 9.3k | today | Apache-2.0 | A | 57 | Claude Code, Codex |