signalbrain AGENTS.md

signalbrain AGENTS.md is an instructions file for Codex, OpenCode from whitestone1121-web/signalbrain. It costs 415 tokens per session, scanned A, original, Apache-2.0.

Repository instructions for reviewing and contributing to SignalBrain, a small project focused on checking whether agent-generated results can be trusted.

In plain words
What is it for?
Use it when reproducing the project demo, checking its command-line checks and reports, or preparing a narrow, verifiable pull request.
Why use it?
They define what to read, how to make reproducible claims, which checks to review, and which unsafe testing activities are prohibited.

Instructions file for CodexOpenCode

Written for Codex and OpenCode: the file is AGENTS.md.

Install

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.

agentmods
npx agentmods add instructions/whitestone1121-web/signalbrain/agents-md
Clone the repo
git clone --depth 1 https://github.com/whitestone1121-web/signalbrain

Made for: Codex, OpenCode.

Wrote 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.

agentmods badge for signalbrain AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/whitestone1121-web/signalbrain/agents-md.svg)](https://agentmods.dev/instructions/whitestone1121-web/signalbrain/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/whitestone1121-web/signalbrain/agents-md"><img src="https://agentmods.dev/badge/instructions/whitestone1121-web/signalbrain/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 415 This file is loaded in full into every session.
When invoked 415 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00415 $0.00415
Opus 5 $0.00208 $0.00208
Sonnet 5 $0.00083 $0.00083
Haiku 4.5 $0.00042 $0.00042

Measured 5d ago against content hash c311f0351894, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

signalbrain 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 5d 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.

AGENTS.md · 52 lines

What it actually says

SignalBrain Agent Instructions

SignalBrain welcomes agent-assisted review and contributions. This repository is small on purpose: the useful work is to test the trust invariant, find unclear edges, and submit reproducible fixes.

First Read

  • README.md for product scope and the integrity boundary.
  • docs/RECEIPT_SPEC.md for the receipt and trust math contract.
  • docs/AGENT_BEACON.md for open review missions and reporting format.
  • SECURITY.md for vulnerability reporting.

Participation Rules

  • Be model-neutral. State the tool or model only when it helps reproduce the finding.
  • Do not submit claims without commands that re-run them.
  • Do not self-grade trust. Reports and PRs should separate what the agent changed from how a reviewer can verify it.
  • Do not use network scans, credential probes, dependency confusion tests, or destructive commands against third-party systems.
  • Keep PRs narrow. One behavior, one fix, one validation trail.

Review Targets

Good autonomous review work includes:

  • Reproducing the demo and quick-start flow on a fresh machine.
  • Checking whether sb check, sb score, sb gate, and sb report fail closed with clear diagnostics.
  • Finding places where docs claim more than the CLI actually verifies.
  • Testing portability across Linux, macOS, Windows, Python versions, and shell environments.
  • Adding small integration examples for other agent stacks.
  • Improving tests where a failure mode is real and reproducible.

PR Contract

Every agent-assisted PR should include:

  • What changed.
  • Why it matters for the integrity invariant.
  • Exact validation commands and observed output.
  • Limitations or remaining uncertainty.
  • A receipt-style claim when the change asserts a measurable improvement.

If the change affects trust math, receipt parsing, scoring, gating, or merged receipt guards, reuse canonical functions instead of local predicates. The report, scorer, and gate must tell the same story.

Changes

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.

  1. 5d ago First seen · 52 lines · 415 tokens per session scan A c311f0351894

Subscribe to this mod's changes

signalbrain AGENTS.md is an instructions file published in the GitHub repository whitestone1121-web/signalbrain (0 stars, last pushed 12d ago), licensed Apache-2.0. It adds 415 tokens to every session, about $0.0021 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.