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/prinova/graphone/agents-mdgit clone --depth 1 https://github.com/PriNova/graphoneWhat 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 | $0.02832 | $0.02832 |
| Opus 5 | $0.01416 | $0.01416 |
| Sonnet 5 | $0.00566 | $0.00566 |
| Haiku 4.5 | $0.00283 | $0.00283 |
Grade D, and why
graphone AGENTS.md scanned grade D with 3 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
| `makensis.exe: No such file` | `sudo apt install nsis` or use `build:windows:exe` | Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
- Requires **bun**: `curl -fsSL https://bun.sh/install | bash` Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- Requires **bun**: `curl -fsSL https://bun.sh/install | bash` How it starts
The opening of the file, as written. The whole thing — 217 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Graphone - Agent Quick Reference
Agent Workflow Guidelines
- Check local code first - Before using GitHub tools (
read_github,search_github, etc.), always explore the local codebase using standard file tools (read,bashwithfind/grep, etc.) - Check pi-mono first for feature implementations - Since graphone is a visual wrapper around pi-mono, look up implementations in local
../pi-monofirst (when available). Many features already exist there and can be mimicked or adapted for desktop UI - Use GitHub tools only when you need to reference external repositories or when explicitly asked about remote code
- Do not track or push
docs/changes - Thedocs/folder and all files under it are local-only in this repo context. Do notgit add/commit/push anything indocs/(including withgit add -f) unless the user explicitly overrides this rule in the current task. - Use title + body commit messages by default - Unless the user explicitly requests otherwise, write commits with a concise title and a descriptive body (not title-only).
- Use robust multiline commit messages - Do not embed literal
\nin a singlegit commit -mstring. Use multiple-mflags (one per paragraph) orgit commit -Fwith a heredoc/file so line breaks render correctly on GitHub. - PAT push hygiene - If pushing with
GITHUB_PAT, pass it via git auth headers/env only; never echo, print, or otherwise expose the token in commands/output. - Default autonomous iteration guidance - For general coding-agent-first execution patterns (tests, TDD/debug loops, logs, traces, artifacts, contracts, smokes, escalation), see
docs/agent-first-autonomous-iteration-playbook.md - Visual verification uses readiness loops, not fixed sleeps - Follow the global workflow
bash -> readiness loop -> take_screenshot -> bash cleanup. The readiness wait must happen inside the startupbashcall because screenshot tools cannot run until that tool call returns. - Do not start and stop dev helpers in parallel - Startup and cleanup are separate sequential tool calls, with
take_screenshotbetween them. Runningstart-dev-and-wait.shandstop-dev.shin parallel is invalid and defeats the workflow. - Prefer project helper scripts for visual checks - For Graphone, use
tooling/scripts/start-dev-and-wait.shandtooling/scripts/stop-dev.shinstead of rewriting startup/cleanup loops inline.
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 · 217 lines · 2,832 tokens per session scan D b49a4f9127be
graphone AGENTS.md is an instructions file published in the GitHub repository PriNova/graphone (88 stars, last pushed 2mo ago), licensed MIT. It adds 2,832 tokens to every session, about $0.0142 per session on Opus 5. A static security scan graded it D with 3 findings (asks for root, downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
dsh-TUI AGENTS.md
AGENTS.md instructions for ccch1mneyyy/dsh-TUI, covering agents.md, 仓库布局, 命令, 上游边界与契约 and 约定与红线.
golembot CLAUDE.md
Instructions for 0xranx/golembot, covering golembot — project conventions for claude code, architecture hard constraints, things you must never do, interface change rules and file responsibility boundaries.
sortie copilot-instructions.md
Instructions for sortie-ai/sortie, covering sortie coding & review standards, 1. layered imports (downward only; violation is critical), 2. concurrency safety, 3. workspace path safety (critical: security boundary) and 4. persistence (sqlite).
copperhead CLAUDE.md
Instructions for copperheadhq/copperhead, covering claude.md, what this repo is, sources of truth, workflow (openspec) and architecture (per spec.md §2).
pi-ask AGENTS.md
Instructions for eko24ive/pi-ask, covering agents.md, project scope, project goals, tech stack and commands.
amp-acp AGENTS.md
Instructions for tao12345666333/amp-acp, covering agents.md, commands, architecture and code style.