What is kosmokrator CLAUDE.md?

A set of Claude Code instructions for KosmoKrator, a terminal-based AI coding agent built with PHP and Symfony that can display text or a terminal interface.

Starting KosmoKrator, choosing its display mode, locating agent and rendering code, and understanding how its tools and subagents fit together.

It gives contributors a map of the project's startup flow, components, tools, and important directories.

kosmokrator CLAUDE.md — Instructions file for Claude Code by OpenCompanyApp: A set of Claude Code instructions for KosmoKrator, a…

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/opencompanyapp/kosmokrator/claude-md

Clone the repo

git clone --depth 1 https://github.com/OpenCompanyApp/kosmokrator

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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/opencompanyapp/kosmokrator/claude-md.svg)](https://agentmods.dev/instructions/opencompanyapp/kosmokrator/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/opencompanyapp/kosmokrator/claude-md"><img src="https://agentmods.dev/badge/instructions/opencompanyapp/kosmokrator/claude-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 1,496 This file is loaded in full into every session.

When invoked 1,496 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 5.1	$0.01496	$0.01496
Opus 5	$0.00748	$0.00748
Sonnet 5	$0.00299	$0.00299
Haiku 4.5	$0.00150	$0.00150

Measured 6d ago against content hash 1d91f856a8fb, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

kosmokrator 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 6d 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 · 107 lines

How it starts

The opening of the file, as written. The whole thing — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.

KosmoKrator

AI coding agent for the terminal. Mythology-themed CLI built with PHP 8.4, Symfony Console, and a dual renderer (TUI/ANSI).

Quick Start

bin/kosmokrator              # Run with auto-detected renderer (TUI if available, ANSI fallback)
bin/kosmokrator --renderer=ansi   # Force ANSI mode
bin/kosmokrator --no-animation    # Skip the animated intro

Architecture

bin/kosmokrator → Kernel → AgentCommand → AgentSessionBuilder → AgentLoop (REPL)
                                            ├── ToolExecutor → tools + PermissionEvaluator
                                            ├── ContextManager → compaction, pruning, system prompt
                                            ├── StuckDetector → headless loop convergence
                                            ├── LLM client (AsyncLlmClient or PrismService)
                                            ├── UIManager → TuiRenderer | AnsiRenderer
                                            ├── ToolRegistry → tools (bash, file_read, file_write, file_edit, grep, glob)
                                            └── SubagentOrchestrator → parallel child agents

Key directories

src/Agent/ — Agent core: AgentLoop (REPL orchestrator), ToolExecutor, ContextManager, StuckDetector, subagent system
src/LLM/ — LLM clients: AsyncLlmClient (Amp HTTP, async), PrismService (Prism PHP, sync), RetryableLlmClient (decorator)
- prism-relay boundary: Provider-specific logic (SSE parsing quirks, stream_options support, usage extraction formats, finish_reason mapping, error normalization, prompt caching strategies) belongs in the opencompany/prism-relay package, NOT in KosmoKrator's src/LLM/. KosmoKrator's LLM layer should only contain agent-level orchestration: retry policy, streaming-to-UI bridging, cancellation handling. Provider capabilities (e.g. supportsStreamUsage()) are declared in prism-relay's ProviderCapabilities and RelayRegistry, consumed via the registry in AsyncLlmClient.
src/UI/ — Rendering layer with split interface hierarchy
- UI/Tui/ — Symfony TUI renderer: TuiRenderer, TuiModalManager (dialogs), TuiAnimationManager (breathing/spinners), SubagentDisplayManager, widgets
- UI/Ansi/ — Pure ANSI fallback: AnsiRenderer, MarkdownToAnsi (with Handler/ for table/list extraction), AnsiTableRenderer
- UI/Diff/ — Unified diff rendering with word-level highlighting
- UI/Theme.php — Shared color palette, tool icons, context bar
- UI/AgentDisplayFormatter.php — Shared agent display utilities (used by both renderers)
- UI/AgentTreeBuilder.php — Builds agent tree from orchestrator stats
src/Tool/ — Tool implementations in Coding/, permission system in Permission/
src/Command/ — AgentCommand (main REPL), SetupCommand, slash commands in Slash/
src/Session/ — SQLite persistence: sessions, messages, memories, settings
src/Task/ — Task tracking system with tool integrations

Read the full file on GitHub · 107 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.

6d ago First seen · 107 lines · 1,496 tokens per session scan A 1d91f856a8fb

Subscribe to this mod's changes

kosmokrator CLAUDE.md is an instructions file published in the GitHub repository OpenCompanyApp/kosmokrator (20 stars, last pushed 5d ago), licensed MIT. It adds 1,496 tokens to every session, about $0.0075 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.

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).

microsoft/vscode · 6,785 tokens

kosmokrator CLAUDE.md

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

KosmoKrator

Quick Start

Architecture

Key directories

What this file has done since we first saw it

Other instructions, from other repositories

vscode buildNext.instructions.md

codex AGENTS.md

vscode oss-third-party-notices.instructions.md

spec-kit AGENTS.md

next.js AGENTS.md

langchain AGENTS.md