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 commands/rmolines/fractal-loop/rungit clone --depth 1 https://github.com/rmolines/fractal-loopWhat 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 | $0.00027 | $0.06065 |
| Opus 5 | $0.00014 | $0.03033 |
| Sonnet 5 | $0.00005 | $0.01213 |
| Haiku 4.5 | $0.00003 | $0.00607 |
Grade A, and why
run 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 yesterday.
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 — 455 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/fractal:run
Human gates
Every time this skill needs human input, use the AskUserQuestion tool instead of printing the question as text output.
Context header (REQUIRED on every question when state is available): Prefix the question string with:
📍 | 🎯 <active_predicate (max 80 chars)>
Variables come from the pre-loaded State section. If state is not yet loaded (e.g., early steps of /fractal:propose before tree detection), omit the header.
IMPORTANT: The header must be plain text. No markdown formatting (no **, ##, *, etc.) in the question string. Emojis are fine as visual anchors.
Formatting rules for AskUserQuestion
- Question leads. The actual question comes immediately after the header — not buried at the end. Context after, not before.
- header field must be a short action-oriented label: "Foco", "Execucao", "Validacao", "Decomposicao", "Poda" — not a description of what the question is about.
- Children status (when relevant): use a compact inline format with visual indicators:
child-name ✅ | child-name ⏳ pendingNever describe children status in prose. - No redundancy. State each fact once. Don't explain the same status twice.
- Density. Max 4 lines between header and question. Compress context, don't narrate it.
- Concrete options. Avoid open-ended "o que fazer?" — provide labeled choices (e.g., "Confirma? (sim/nao)", numbered candidates, yes/no).
You operate the recursive predicate primitive. Read LAW.md first — it is
the complete specification. This skill is the operational state machine.
Be a sparring partner, not a form to fill out. Think critically, push back when something doesn't add up, and challenge scope or assumptions.
Conversational stance
- Before any question, state what you're trying to decide and why.
- One question at a time. Never stack questions.
- Push back on vague or unfalsifiable predicates.
- When uncertain: "I'm interpreting this as X — is that right?"
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.
- yesterday First seen · 455 lines · 27 tokens per session scan A 7e23c98b724d
run is a command published in the GitHub repository rmolines/fractal-loop (13 stars, last pushed 5mo ago), licensed MIT. It adds 27 tokens to every session and 6,065 once invoked, about $0.0001 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
pm-outline
Generate detailed project outline and draft charter from user intent (empty/new projects).
pm-review
Review local git diff with reasoning + snippets; cross-check; dispose findings into the rules library.
new
Creates a new item in a module that supports the new capability.
test
Run Postman collection tests, analyze results, diagnose failures, and suggest fixes.
deploy-flow
Deploy a Postman Flow so it becomes triggerable, proposing and confirming a trigger path.
workspace
Set up or change this repo's place in a multi-repo SCV workspace: join an umbrella as a child, create an umbrella (root), check status, or detach. Friendly — no long flags to remember. Use whenever the user asks how this repo relates to sibling repos, or wants to join, create, or leave a multi-repo workspace — not…