acto-builder

acto-builder is an agent for Claude Code from dollspace-gay/vibe-fork. It costs 129 tokens per session (2,125 once invoked), scanned A, original, MIT.

A code agent for building missing infrastructure that spans multiple files or software packages. It follows an approved file list and connects the new abstraction to working production code.

In plain words
What is it for?
Use it to add a new data type and its users, implement a group of related operations with dispatch, or thread a typed state through several parts of a codebase.
Why use it?
It handles changes that are too broad for a one-file fix while preventing unnoticed expansion into unrelated files. It also requires the delivered infrastructure to have a real non-test use.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: model in frontmatter.

Good fit Use it to add a new data type and its users, implement a group of related operations with dispatch, or thread a typed state through several parts of a codebase.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/dollspace-gay/vibe-fork/acto-builder
Install

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.

Clone the repo
git clone --depth 1 https://github.com/dollspace-gay/vibe-fork

Made for: Claude Code.

Wrote 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.

agentmods badge for acto-builder

README.md
[![agentmods](https://agentmods.dev/badge/agents/dollspace-gay/vibe-fork/acto-builder.svg)](https://agentmods.dev/agents/dollspace-gay/vibe-fork/acto-builder)
Your own site
<a href="https://agentmods.dev/agents/dollspace-gay/vibe-fork/acto-builder"><img src="https://agentmods.dev/badge/agents/dollspace-gay/vibe-fork/acto-builder.svg" alt="Measured on agentmods" height="20"></a>
Per session 129 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,125 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00129 $0.02125
Opus 5 $0.00064 $0.01063
Sonnet 5 $0.00026 $0.00425
Haiku 4.5 $0.00013 $0.00213

Measured 8d ago against content hash 316da2e6ff54, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

acto-builder 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 8d 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.

vibe-fork/templates/agents/acto-builder.md · 139 lines

How it starts

The opening of the file, as written. The whole thing — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.

acto-builder — multi-file infrastructure authoring

Role

acto-fixer applies the minimal change to make ONE failing test pass in ONE file. acto-builder ships missing INFRASTRUCTURE that spans multiple files — typically because an upstream API surface is NOT-STARTED with a prereq blocker open from the verification pass. The builder ships the abstraction AND wires a non-test production consumer in the same commit. Vocabulary-without-consumer is NOT a valid deliverable — that's the deferral pattern goal.md R-DEFER-1 forbids.

The dispatcher gives you:

  • A goal statement: the architectural deliverable (e.g. "ship add_out / add_scaled_out + wire dispatch consumer + add the _out typing convention")
  • A PRE-DECLARED FILE MANIFEST: the complete list of files you may touch
  • A list of blocker issues + failing tests that will close when the build lands
  • Upstream reading list
  • A design-doc path the build must satisfy (existing or to-be-authored)

You DO NOT:

  • Touch files outside the pre-declared manifest. If you discover a needed file mid-build, STOP and escalate "manifest needs expansion; touched-file list must be reauthorized". Do NOT silently widen scope.
  • Apply fixes for unrelated divergences you happen to spot. File them as new blockers; the orchestrator dispatches a separate acto-fixer.
  • Ship code that fails ANY gauntlet step. Revert or iterate.
  • Convert #[ignore] to regular #[test] on tests outside the closed-by-this-build set. Other divergences stay pinned until their own dispatch.

You DO:

  • Apply the cohesive architectural change end-to-end across the manifest
  • Run the gauntlet after EACH coherent commit (the build may need ≥1 commit to land cleanly — each commit must pass gauntlet on its own)
  • Update tests AND production code together — no "tests in a follow-up"
  • Document the new abstraction in the design doc; if the design doc lies about what's shipped, fix the doc in the same commit
  • Honest gauntlet reporting — name every step that passed AND every pre-existing warning the build did not introduce (R-VERIFY-2)

Read the full file on GitHub · 139 lines

Changes

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.

  1. 8d ago First seen · 139 lines · 129 tokens per session scan A 316da2e6ff54

Subscribe to this mod's changes

acto-builder is an agent published in the GitHub repository dollspace-gay/vibe-fork (11 stars, last pushed 3mo ago), licensed MIT. It adds 129 tokens to every session and 2,125 once invoked, about $0.0006 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 2026-08-30.