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/hyperion-gpu/proofflow-v0.1/agents-mdgit clone --depth 1 https://github.com/Hyperion-GPU/ProofFlow-v0.1Wrote 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/hyperion-gpu/proofflow-v0.1/agents-md)<a href="https://agentmods.dev/instructions/hyperion-gpu/proofflow-v0.1/agents-md"><img src="https://agentmods.dev/badge/instructions/hyperion-gpu/proofflow-v0.1/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 | $0.00484 | $0.00484 |
| Opus 5 | $0.00242 | $0.00242 |
| Sonnet 5 | $0.00097 | $0.00097 |
| Haiku 4.5 | $0.00048 | $0.00048 |
Grade A, and why
ProofFlow-v0.1 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 4d 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 — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ProofFlow Agent Rules
These rules apply to coding agents working in this repository.
Workflow
- Communicate and deliver in Simplified Chinese unless the user asks otherwise.
- Start with a quick audit of the current files and constraints before changing code or docs.
- Keep changes small, complete, and directly tied to the requested milestone.
- Prefer root-cause fixes over broad rewrites.
- Do not add cloud services, remote sync, telemetry, or Docker unless the user explicitly changes the v0.1 scope.
Evidence
- AI or heuristic claims must cite evidence: file paths, command output, test results, screenshots, or user-provided source material.
- Code review findings must explain the impact and point to the relevant file or symbol.
- If evidence is missing, label the statement as an assumption.
Safety
- Destructive file actions require three parts before execution:
- dry-run command or preview,
- explicit user approval,
- undo or recovery path.
- Do not delete user data, local databases, generated proof packets, or uploaded artifacts without approval.
- Use Windows-friendly commands and quote paths that may contain spaces.
Architecture
- Backend code should stay modular: routers handle HTTP shape, services handle workflow logic, and database helpers own SQLite access.
- Frontend code should stay simple until the MVP workflows are proven.
- Keep localhost as the default trust boundary.
- Do not implement business features in scaffolding commits unless the milestone specifically calls for them.
ProofFlow product invariants
- No Case, no workflow.
- No Evidence, no trusted Claim.
- No Preview, no Action.
- No Undo, no destructive Action.
- No Test, no accepted code workflow.
- No Source, no Artifact.
Managed backup / restore invariants
- No Manifest, no Backup.
- No Verify, no trusted Backup.
- No Preview, no Restore.
- No Pre-restore Backup, no destructive Restore.
- No Hash Match, no Restore.
- No Source Version, no Restore Trust.
- No Restore to live DB in foundation phase.
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.
- 4d ago First seen · 70 lines · 484 tokens per session scan A c05dc4be8fd4
ProofFlow-v0.1 AGENTS.md is an instructions file published in the GitHub repository Hyperion-GPU/ProofFlow-v0.1 (48 stars, last pushed 27d ago), licensed MIT. It adds 484 tokens to every session, about $0.0024 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.
Other instructions, from other repositories
Sponsio CLAUDE.md
Claude Code instructions for SponsioLabs/Sponsio, covering agent guide for sponsio, what sponsio is, positioning, what to read first and repository map.
jentic-one GEMINI.md
Instructions for jentic/jentic-one: Otherwise, read AGENTS.md — this repo's canonical agent guidance.
hig-doctor AGENTS.md
AGENTS.md instructions for raintree-technology/hig-doctor: HIG Doctor combines open-source audit tooling with an attributed snapshot of Apple's Human Interface Guidelines.
briefloop AGENTS.md
AGENTS.md instructions for Stahl-G/briefloop, covering agents.md, purpose, instruction scope, environment separation and context mode.
obsigna python.instructions.md
Instructions for agent-receipts/obsigna, a project described as: Agent Receipts — cryptographically signed audit trails for AI agent actions. Protocol spec, SDKs (Go, TypeScript, Python), and MCP proxy.
goshipit CLAUDE.md
Instructions for Capta1nRaj/goshipit, covering goshipit - project instructions, sync rule (enforce after every change), how to count checks, git commits and package.json description format.