Borrowing it
Nothing to install: this file belongs to goldsky-io/goldsky-agent. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/goldsky-io/goldsky-agent/main/AGENTS.mdgit clone --depth 1 https://github.com/goldsky-io/goldsky-agentWrote 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/goldsky-io/goldsky-agent/agents-md)<a href="https://agentmods.dev/instructions/goldsky-io/goldsky-agent/agents-md"><img src="https://agentmods.dev/badge/instructions/goldsky-io/goldsky-agent/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/goldsky-io/goldsky-agent/agents-md"><img src="https://agentmods.dev/badge/instructions/goldsky-io/goldsky-agent/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 | $0.01553 | $0.01553 |
| Opus 5.5 | $0.00621 | $0.00621 |
| Sonnet 5.5 | $0.00311 | $0.00311 |
| Haiku 4.5 | $0.00155 | $0.00155 |
Grade A, and why
goldsky-agent 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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — Goldsky Agent
Guide for AI coding agents (Devin, Codex, Cursor, Amp, Claude Code, and anything else that reads AGENTS.md). This repo is a skill pack for building, deploying, and debugging on Goldsky — Turbo pipelines, Mirror pipelines, Subgraphs, Edge RPC, and Feeds. Each skill is a self-contained folder of step-by-step instructions plus reference material. Compose skills, which send onchain transactions, live in goldsky-io/goldsky-compose-agent.
How to use these skills
Some hosts (Claude Code, Cursor) auto-discover skills and trigger them from each skill's description. If your runtime has no "skills" feature, use them manually — they are just Markdown:
- Identify the task (build a pipeline, fix a broken one, write a transform, deploy/repair a subgraph…).
- Open the matching
skills/<name>/SKILL.md(routing table below; full index inSKILLS.md). - Follow it. A
SKILL.mdpoints into deeper files underskills/<name>/references/andskills/<name>/templates/— read those when it tells you to. - For reference lookups (CLI flags, YAML fields, dataset names, error codes), query the Goldsky docs MCP at
https://docs.goldsky.com/mcprather than guessing. Every published doc is also inhttps://docs.goldsky.com/llms.txt.
Each SKILL.md opens with a description: in its frontmatter stating what it covers and when to use it — scan those to route.
Skill routing
| If the task is… | Use skill |
|---|---|
| Build/deploy a new Turbo pipeline (chain → dataset → transforms → sink) | turbo-builder |
| A Turbo pipeline is broken / not getting data / output looks wrong | turbo-doctor |
| Turbo YAML syntax, sources/sinks, architecture & sizing decisions | turbo-pipelines |
| Write a SQL / TypeScript / dynamic-table transform (incl. EVM & Solana decoding) | turbo-transforms |
| Pause / restart / delete a pipeline; interpret a pipeline error | turbo-operations |
| Sync a subgraph entity source to a DB; Mirror vs Turbo | mirror |
| A Mirror pipeline is failing / stuck / terminated | mirror-doctor |
| Author / build / deploy a subgraph; schema, mappings, manifest | subgraph-builder |
| A subgraph stopped syncing / won't deploy / errors | subgraph-doctor |
| Migrate a subgraph off The Graph | subgraph-migrate |
| Get a fast, reliable managed RPC endpoint; RPC error codes | edge |
Cut the bill from an RPC provider you already pay for; cache hit rates, x-cache |
boost |
| Look up one wallet's balances or transfers, token metadata and USD prices, a wallet's Polymarket activity or positions, or a block by number, hash or timestamp, over the Feeds REST API (formerly Edge Data / Data APIs) | feeds |
| Find the right dataset name / chain prefix | datasets |
| Store credentials for a sink (Postgres, ClickHouse, Kafka…) | secrets |
Install the CLI / log in / switch projects / fix unauthorized |
auth-setup |
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 Changed · +27 tokens per session 1dbc58ac9482
- 3d ago Changed · +8 tokens per session e4262e08a352
- 4d ago Changed · -3 lines · -38 tokens per session 49df8caa2f01
- 28d ago Changed · +1 lines · +23 tokens per session 6a631bc493d2
- 1mo ago Changed · -2 lines · +15 tokens per session 553a7485ac16
- 1mo ago Changed · +1 lines · +26 tokens per session 7cb8478f30d3
- 2mo ago First seen · 79 lines · 1,492 tokens per session scan A a62da6900131
goldsky-agent AGENTS.md is an instructions file published in the GitHub repository goldsky-io/goldsky-agent (12 stars, last pushed yesterday), licensed MIT. It adds 1,553 tokens to every session, about $0.0062 per session on Opus 5.5. 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
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.