Borrowing it
Nothing to install: this file belongs to NexgenSystemsMX/scopegate. 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/NexgenSystemsMX/scopegate/master/AGENTS.mdgit clone --depth 1 https://github.com/NexgenSystemsMX/scopegateWrote 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/nexgensystemsmx/scopegate/agents-md)<a href="https://agentmods.dev/instructions/nexgensystemsmx/scopegate/agents-md"><img src="https://agentmods.dev/badge/instructions/nexgensystemsmx/scopegate/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/nexgensystemsmx/scopegate/agents-md"><img src="https://agentmods.dev/badge/instructions/nexgensystemsmx/scopegate/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 | $0.02983 | $0.02983 |
| Opus 5.5 | $0.01193 | $0.01193 |
| Sonnet 5.5 | $0.00597 | $0.00597 |
| Haiku 4.5 | $0.00298 | $0.00298 |
Grade A, and why
scopegate 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 — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — ScopeGate repo guide for coding agents
ScopeGate is an ephemeral-credentials gateway for coding agents (Claude Code, Kimi Code, Cursor, OpenCode, or any MCP harness). The agent never holds secrets — it holds short-lived, minimum-scope capabilities minted by a policy engine. Secrets live in an AES-256-GCM encrypted vault and are injected only at the outbound hop.
If you want to USE ScopeGate
Do not read the source. The complete, agent-executable usage documentation lives in docs/agents/README.md — eight guides covering self-install, the operating protocol, the MCP tools reference, connectors, policies, self-repair, security rules, and long tasks. Start at guide 01 and follow it literally.
A drop-in protocol summary also ships as SKILL.md (usable as a
skills-dir entry or a CLAUDE.md / AGENTS.md fragment).
If you are DEVELOPING ScopeGate
Requirements: Node.js ≥ 20, npm 10. No native modules, no global services.
Repository: NexgenSystemsMX/scopegate, default branch master.
Surface facts (verified 2026-08): the agent-facing MCP server exposes 21
scopegate_* tools (frozen list in src/gateway/tools.ts); the bundled
bridges expose huly 16 / railway 7 / cloudflare 8 / google 7 tools.
Deployment (Railway)
The public product deploys from the image ghcr.io/luisrosasx/scopegate
(built by docker-ghcr.yml on push to master) into the Railway project
scopegate: service scopegate (gateway with SCOPEGATE_SEED_DEMO=1 —
100% fake demo data) and scopegate-cloud (ScopeGate Cloud: landing, panel
and /v1 API) at scopegate.io. The Nexgen
ecosystem's production gateways do not deploy from this repo — they run
as services of the Railway project kimi-tag. Full map:
docs/ecosistema.md.
Layout (src/)
cli.ts # entry: init | start | secret | status | audit | auth | vaultd | vault |
# rollback | approve | deny | policies | cloud | git-credential |
# inject | honeytoken
api.ts # M7: embeddable library API (createGatewayServer) — dist/api.js + types
testkit/ # M7: scopegate/testkit — fake upstream + bootFakeGateway (consumer tests)
commands/ # init, secret, oauth-login, vaultd, vault-rotate, approvals/policies CLI,
# git-credential, inject
config/ # paths + scopegate.yaml loader (NEVER holds secrets)
gateway/ # MCP server, policy enforcement, upstream proxy + injection + self-heal,
# scopegate_* tools (transports: stdio, http, openapi)
harness/ # adapters: claude-code, kimi-code, cursor, opencode, mcp-json
vault/ # AES-256-GCM store, master-key backends, vaultd daemon + IPC
policy/ # engine (when: arg guards, vault:inject default escalation), grants,
# limits, rate limit, redact, approvals
inject/ # M10: governed secret materialization into files (atomic 0600 + sidecar)
minter/ # token minter + providers/ (jwt, github-app, aws-sts, huly, google-sa)
oauth/ # refresh daemon, device-code
audit/ # hash-chained JSONL, Ed25519 signing, verify, index
upstreams/ # native connectors: huly, railway, cloudflare, google + openapi importer
cloud/ # optional multi-tenant control plane (metadata-only; FileStore|PostgresStore)
attestation/ honeytoken/ notify/ registry/ telemetry/
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 Changed · +9 lines · +345 tokens per session d3cacf30e794
- 1mo ago First seen · 135 lines · 2,638 tokens per session scan A d66f520d7fb2
scopegate AGENTS.md is an instructions file published in the GitHub repository NexgenSystemsMX/scopegate (1 stars, last pushed 7d ago), licensed Apache-2.0. It adds 2,983 tokens to every session, about $0.0119 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 2026-08-31.
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).
vscode model-prompts.instructions.md
Model-specific prompt authoring and registry guidelines.