Borrowing it
Nothing to install: this file belongs to akougkas/helios-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/akougkas/helios-mcp/main/CLAUDE.mdgit clone --depth 1 https://github.com/akougkas/helios-mcpWrote 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.
[](https://agentmods.dev/instructions/akougkas/helios-mcp/claude-md)<a href="https://agentmods.dev/instructions/akougkas/helios-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/akougkas/helios-mcp/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/akougkas/helios-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/akougkas/helios-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.02305 | $0.02305 |
| Opus 5.5 | $0.00922 | $0.00922 |
| Sonnet 5.5 | $0.00461 | $0.00461 |
| Haiku 4.5 | $0.00231 | $0.00231 |
Grade A, and why
helios-mcp 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 1mo 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 — 138 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.
Helios
The behavioral genome for AI agents. Observes actual behavior, detects drift via a Dirichlet posterior and Jensen-Shannon divergence, negotiates profile updates with the human. Every change is git-committed. Ships as a Claude Code plugin (hooks + skills + MCP).
Not a RAG system, not a preference store, not a static personality file.
Commands
uv sync # install deps
uv run pytest tests/ -q # all tests (598)
uv run pytest tests/test_distribution.py # single file
uv run pytest -k "js_is_symmetric" # by name
uv run pytest --cov=src/helios_mcp # coverage
uv run ruff check src/ tests/ # lint
uv run mypy src/helios_mcp/ # types
uv build --wheel # build
uv run helios-mcp # start MCP server
uv run helios-mcp init # onboard: import CLAUDE.md + output style, set default persona
uv run helios-mcp status # show profiles
uv run helios-mcp negotiate <persona> # drift report
Architecture
The Cycle
observe_interaction → get_drift_report → negotiate_update → git commit
↑ |
└────────────── next session ────────────┘
The plugin's hooks capture derived features only (never prompt or tool content) and feed a per-turn heuristic classifier plus, at session end, a batch Haiku labeler. Both write soft labels over 9 behavioral dimensions into an append-only ledger. The estimator turns that ledger into a Dirichlet posterior per dimension and measures Jensen-Shannon divergence against the declared profile; see "Key Thresholds" below for the exact math. Human accepts or rejects each proposal. Accepted changes are YAML-written and git-committed.
Identity Hierarchy
species (base/identity.yaml) weight = base_importance / (specialization_level²)
└→ domain (personas/developer.yaml) M(x) = w·P_parent(x) + (1-w)·P_child(x)
└→ user (personas/dev_user.yaml)
└→ session (temporary/dev.yaml)
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.
- 1mo ago Changed · -178 lines · -322 tokens per session 9417dbaf78b1
- 2mo ago First seen · 316 lines · 2,627 tokens per session scan A 9cc6a0ffb990
helios-mcp CLAUDE.md is an instructions file published in the GitHub repository akougkas/helios-mcp (2 stars, last pushed 1mo ago), licensed MIT. It adds 2,305 tokens to every session, about $0.0092 per session on Opus 5.5. 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.
helios-mcp CLAUDE.md compared with instructions from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| helios-mcp CLAUDE.mdakougkas/helios-mcp | Instructions file | 2 | 1mo ago | MIT | A | 2,305 | Claude Code |
| vscode buildNext.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 9,719 | GitHub Copilot |
| next.js AGENTS.mdvercel/next.js | Instructions file | 143k | yesterday | MIT | A | 7,461 | Codex, OpenCode |
| langchain AGENTS.mdlangchain-ai/langchain | Instructions file | 149k | yesterday | MIT | A | 4,680 | Codex, OpenCode |
| vscode oss-third-party-notices.instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 5,516 | GitHub Copilot |
| vscode copilot-instructions.mdmicrosoft/vscode | Instructions file | 195k | yesterday | MIT | A | 3,093 | GitHub Copilot |
| deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness | Instructions file | 274k | 2d ago | MIT | A | 4,313 | Codex, OpenCode |