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 instructions/acquiredl/acu/claude-mdgit clone --depth 1 https://github.com/Acquiredl/AcuWhat 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 | $0.01228 | $0.01228 |
| Opus 5 | $0.00614 | $0.00614 |
| Sonnet 5 | $0.00246 | $0.00246 |
| Haiku 4.5 | $0.00123 | $0.00123 |
Grade A, and why
Acu CLAUDE.md 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 — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Acu
A framework for building LLM-driven pipelines where the filesystem defines the program. You create directories for stages, CLAUDE.md files for behavior, and quality gates for reliability. The LLM reads this structure and executes it — templates let you replicate the pattern across new domains.
Because LLMs are probabilistic, Acu treats quality gates as a first-class concern: every stage transition must pass a deterministic check before work advances. This is what makes the output reliable despite the executor being non-deterministic.
Anchor Metaphor
Acu is an office. Each pipeline is a department with its own charter (its CLAUDE.md). Each stage is a desk inside the department. Between desks there is an inspection point — a quality gate — that signs off on work before it advances. A work unit is a project moving from desk to desk through the department. The department's final deliverable is the project's report.
The Orchestrator is Operations — the only role with cross-department visibility. It routes work, reviews outputs against standards, and surfaces framework-level improvements back into how departments run. Departments stay isolated from each other; Operations is the one seam.
This is the mental model. The rest of this document — the mechanical names (pipeline, stage, gate, unit, deliverable, Orchestrator) and the structural rules — are how the office runs in practice.
Architectural Principles
Acu's architecture is organized around isolation, gates, and structure:
- Isolation — Each pipeline runs independently with its own stages, gates, and context. No pipeline can reach into another. The Orchestrator is the only component with cross-pipeline visibility.
- Two-layer gates — Every stage transition passes a deterministic structural check (file existence, word counts, section headers, cross-references) followed, where configured, by an LLM-based semantic evaluation across three tiers (stage, pipeline, system). The structural layer is binary; the semantic layer implements the Evaluator-Optimizer loop with bounded retries. Nothing advances without passing both.
- Structure as schema — Templates enforce directory layout and file requirements. CLAUDE.md files scope what context gets loaded. Validation is structural, not semantic.
- Audit trail — Every transition is logged with session ID and SHA256.
syslog.shaggregates across pipelines. - Threat model — Pipeline isolation, input validation, and defined attack surfaces. See
THREAT-MODEL.md.
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 · 70 lines · 1,228 tokens per session scan A b21b11ae05b6
Acu CLAUDE.md is an instructions file published in the GitHub repository Acquiredl/Acu (2 stars, last pushed 24d ago), licensed MIT. It adds 1,228 tokens to every session, about $0.0061 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 instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.