What is agenture-loop CLAUDE.md?

A project guide for Agenture Loop, a system of commands and helper agents for defining, designing, planning, implementing, and validating software work. It documents the current architecture, task escalation, quality checks, and documentation updates.

Use it to route work through the project’s definition, design, planning, implementation, validation, and documentation-sync commands.

It tells a coding agent which command or helper should handle each stage and when implementation must pause because the design is incomplete.

agenture-loop CLAUDE.md — Instructions file by AgentureHQ

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.

agentmods

npx agentmods add instructions/agenturehq/agenture-loop/claude-md

Clone the repo

git clone --depth 1 https://github.com/AgentureHQ/agenture-loop

Per session 1,963 This file is loaded in full into every session.

When invoked 1,963 The same file — it is already loaded in full.

Security scan A 0 findings. Scan, not verified.

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.

Model	Per session	Once invoked
Fable 5	$0.01963	$0.01963
Opus 5	$0.00981	$0.00981
Sonnet 5	$0.00393	$0.00393
Haiku 4.5	$0.00196	$0.00196

Measured 2d ago against content hash a2a8d73fbc48, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agenture-loop 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.

CLAUDE.md · 117 lines

How it starts

The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Foundational rules to always follow

@.claude/rules/first-principles.md

Architecture rework complete

Epic agentic_sdlc_rework has shipped all six features:

unified_skills_and_cleanup — verb-noun skill surface /agn:define|design|plan|implement|validate <level> is live.
rules_split_and_new_files — composition rules in rules/task-composition.md; persistence in taskman.sh help; role-specific rule files rules/qa.md and rules/doc-maintenance.md.
planner_subagent — Planner sub-agent at plugins/agn/agents/planner.md; /agn:define, /agn:design, /agn:plan delegate composition to it.
task_escalation_protocol — /agn:implement task detects design gaps, writes a gap-log to tasks/gaps/, halts with a routing message, supports resume.
qa_subagent_and_validation — QA sub-agent at plugins/agn/agents/qa.md; /agn:validate feature|epic|product delegate to it; /agn:validate task runs lightweight gates in the main session.
docsync_close_hook — taskman.sh close actions append to tasks/docs-sync-queue.txt; /agn:docs-sync processes the queue, reviews upstream docs for drift per rules/doc-maintenance.md, proposes diffs.

What this repo is

A Claude Code plugin marketplace (agenture) that ships one plugin today: agn — an agentic SDLC loop. There is no application code, no build, and no test framework. The deliverables are:

.claude-plugin/marketplace.json — marketplace manifest
plugins/<name>/ — self-contained plugin directories (skills, rules, scripts, optional agents/commands/hooks/.mcp.json)
docs/ — product docs for the marketplace itself
tasks/ — this repo's own SDLC tracking, dogfooded through agn

Dogfooding pattern (important)

The repo runs its own plugin via symlinks under .claude/:

.claude/skills -> ../plugins/agn/skills
.claude/rules  -> ../plugins/agn/rules

Read the full file on GitHub · 117 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.

2d ago First seen · 117 lines · 1,963 tokens per session scan A a2a8d73fbc48

Subscribe to this mod's changes

agenture-loop CLAUDE.md is an instructions file published in the GitHub repository AgentureHQ/agenture-loop (2 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 1,963 tokens to every session, about $0.0098 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.

Building agents Planning ai-coding claude-code-plugin

agenture-loop CLAUDE.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

CLAUDE.md

Foundational rules to always follow

Architecture rework complete

What this repo is

Dogfooding pattern (important)

What this file has done since we first saw it

Other instructions, from other repositories

project-os-for-codex AGENTS.md

AmbyKit CLAUDE.md

sugar AGENTS.md

nextjs-skills CLAUDE.md

gandalf CLAUDE.md

symbiote CLAUDE.md