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/trentisiete/endy/agents-mdgit clone --depth 1 https://github.com/trentisiete/endyWrote 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/trentisiete/endy/agents-md)<a href="https://agentmods.dev/instructions/trentisiete/endy/agents-md"><img src="https://agentmods.dev/badge/instructions/trentisiete/endy/agents-md.svg" alt="Measured on agentmods" 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 | $0.03029 | $0.03029 |
| Opus 5 | $0.01515 | $0.01515 |
| Sonnet 5 | $0.00606 | $0.00606 |
| Haiku 4.5 | $0.00303 | $0.00303 |
Grade A, and why
endy 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 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.
How it starts
The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
endy — multi-agent coding stack
You're running inside endy, a hybrid bash + skill stack that wires several coding-agent CLIs together. This file is loaded by every supported agent (Codex, OpenCode, CommandCode/cmd, and — when re-enabled — Claude Code) via their global AGENTS.md mechanism. Hermes loads its own context via ~/.hermes/SOUL.md separately.
Read it once at the start of a session and keep it in mind. Don't quote it back at the user.
Who's in the stack
| Agent | Strength | When to pick it |
|---|---|---|
Codex (codex) |
Long-context planning, GitHub-Copilot-backed gpt-5.5 with xhigh reasoning | Default orchestrator. Use Codex itself unless the task fits a specialist. |
OpenCode (opencode) |
Fast multi-model worker, default agent "build" runs on big-pickle | Mechanical refactors, test-writing, codemods. Cheap and fast. |
CommandCode (cmd) |
"Taste-1" model — purpose-built for code aesthetics | "Does this read well / match the codebase's idioms?" Nothing else is this good for that. |
Hermes (hermes) |
Tool-calling-tuned models from Nous Research; rich plugin/MCP/skill ecosystem | Open-ended agentic work, when you want a specific Nous model, or when you want a parallel orchestrator independent of Codex. |
Gemini CLI (gemini) |
Google Gemini, generous free daily quota | Free-tier worker. Use for routine refactors, summaries, exploratory work — keep paid agents for the parts that need them. |
Delegation primitives
Three ways to hand work off — pick deliberately, the differences matter:
- In-Codex personas (
.codex/agents/architect.toml,reviewer.toml,researcher.toml). The nativemulti_agenttoolset spawns these inside Codex itself — no extra processes. Cheap and clean for orchestrator-side roles. - Short bash inline — for blocking calls that finish in a few minutes:
opencode run [--agent <persona>] "<prompt>"cmd --skip-onboarding --trust [--yolo] -p "<prompt>"hermes chat -Q --accept-hooks [--skills <name>] -q "<prompt>"gemini [--model <m>] [--yolo] -p "<prompt>"The skillendy-delegatecovers when to use a persona vs ad-hoc inline.
- Long detached runs — for unsupervised work that may take 5+ minutes and needs full permissions OK:
endy spawn <agent> [--persona <name>] -- "<prompt>"- Returns a TASK_ID immediately; agent runs in a fresh tmux window.
- Track with
endy watch list/endy watch log <id>.
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 · 128 lines · 3,029 tokens per session scan A 9e30948bf690
endy AGENTS.md is an instructions file published in the GitHub repository trentisiete/endy (7 stars, last pushed 3mo ago), licensed MIT. It adds 3,029 tokens to every session, about $0.0151 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
hames CLAUDE.md
Claude Code instructions for baek-labs/hames, a project described as: Workspace-first orchestration for multi-model AI workflows. Frozen reference implementation.
hames AGENTS.md
AGENTS.md instructions for baek-labs/hames: This repository builds the Hames plugin for Codex and Claude Code.
tmux-rc AGENTS.md
AGENTS.md instructions for querystory/tmux-rc, covering agents.md — working conventions for tmux-rc, testing changes on the live dev instance, develop in worktrees, off main, review before merge and ui iconography: no emoji — inline lucide icons.
OctoOffice AGENTS.md
AGENTS.md instructions for Chepko932/OctoOffice, covering octooffice orchestration rules, language rule, default skill bootstrap, workflow orchestration (additive) and 1) plan node default.
cmd CLAUDE.md
Claude Code instructions for AUTMZR/cmd, covering claude.md, 0. говори простым языком, 1. think before coding, 2. simplicity first and 3. surgical changes.
agents CLAUDE.md
Instructions for eloylp/agents, covering claude.md, project overview, directory structure, config model and build & run.