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/evoelsewhere/evoflux/agents-mdgit clone --depth 1 https://github.com/evoelsewhere/evofluxWhat 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 | $0.03398 | $0.03398 |
| Opus 5 | $0.01699 | $0.01699 |
| Sonnet 5 | $0.00680 | $0.00680 |
| Haiku 4.5 | $0.00340 | $0.00340 |
Grade A, and why
evoflux AGENTS.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 — 394 lines — stays where its author put it; the contents beside it link to each section on GitHub.
EvoFlux — Agent Instructions
EvoFlux is a local-first desktop workspace for Work and Coding agent teams. The production application is a Tauri shell embedding a React UI and supervising a local FastAPI sidecar.
Instruction precedence
This file applies to the whole repository. Before changing a file, also read the
nearest nested AGENTS.md; nested instructions add to or override this root
contract for their directory.
Scoped instruction files currently live under:
app/, includingapp/agent/,app/api/, andapp/services/;web/andweb/src/;desktop/anddesktop/src-tauri/;seed/;scripts/.
Repository map
app/ Python sidecar: API, agents, services, persistence, automation
web/ React/TypeScript interface embedded by Tauri
desktop/ Rust/Tauri shell and native packaging
seed/ First-install agent and configuration templates
scripts/ Development, build, validation, and release utilities
tests/ Python backend, integration, CLI, and packaging tests
documents/ Single documentation root and EASD repository data
Start with:
documents/README.mdfor documentation navigation;documents/features/README.mdfor implemented features and code ownership;documents/architecture/system-overview.mdfor process and trust boundaries;documents/reference/repository-map.mdfor the detailed source map.
Architecture boundaries
- Keep FastAPI routes thin. Durable behavior belongs in
app/services/,app/agent/,app/workflow/,app/scheduler/, orapp/core/. - Keep provider-specific payloads behind
app/agent/providers/; generic API and team schemas must remain provider-neutral. - Use TanStack Query for frontend server state, Zustand for live/client state, and shared shell/workbench primitives for application chrome.
- Tauri owns native lifecycle, capabilities, browser integration, and packaging; the Python sidecar owns agent policy, persistence, and workspace authorization.
- The application database stores product state. Repository code indexes remain cache-local per-repository databases and must not move into application tables.
- Work mode uses session workspaces. Coding mode may access only repositories authorized by the active workspace or Coding project.
- Tools, MCP, plugins, browser content, imported documents, and remembered text remain subject to explicit trust, permission, sandbox, and untrusted-data boundaries.
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 · 394 lines · 3,398 tokens per session scan A 1634b8f59e58
evoflux AGENTS.md is an instructions file published in the GitHub repository evoelsewhere/evoflux (5 stars, last pushed 6d ago), licensed Apache-2.0. It adds 3,398 tokens to every session, about $0.0170 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
vibepod-cli AGENTS.md
Instructions for VibePod/vibepod-cli, covering project agent memory, adding a supported agent, tests and maintaining this file.
ankaloop AGENTS.md
AGENTS.md instructions for tao12345666333/ankaloop, covering ankaloop project rules, project overview, code style, architecture patterns and agent behavior.
SztuCode CLAUDE.md
Instructions for rojim666/SztuCode, covering claude.md, commands and architecture.
adeptability AGENTS.md
Instructions for itaywol/adeptability, a project described as: Cross-harness AI skill portability CLI. Author an agent skill once, sync it into Claude Code, Cursor, Copilot, Codex, OpenCode & 45+ AI coding agents. Safety scanner, content-hash drift detection. A package manager / dotfiles for AI coding agent skills.
agent-os CLAUDE.md
Instructions for earthwalker17/agent-os, covering claude.md, 1. project mission, 2. core architecture principles, 3. agent roles and main agent (orchestrator).
Warden-AI CLAUDE.md
Claude Code instructions for rynald0cst0ltziam/Warden-AI, covering warden — context governance and verification layer, session start (important — do this first), layer 1: before starting work — context selection, layer 2: during work — tool output pruning and enforcement hooks (automatic).