Borrowing it
Nothing to install: this file belongs to alifanov/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/alifanov/scopegate/master/CLAUDE.mdgit clone --depth 1 https://github.com/alifanov/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/alifanov/scopegate/claude-md)<a href="https://agentmods.dev/instructions/alifanov/scopegate/claude-md"><img src="https://agentmods.dev/badge/instructions/alifanov/scopegate/claude-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/alifanov/scopegate/claude-md"><img src="https://agentmods.dev/badge/instructions/alifanov/scopegate/claude-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 | $0.07477 | $0.07477 |
| Opus 5 | $0.03739 | $0.03739 |
| Sonnet 5 | $0.01495 | $0.01495 |
| Haiku 4.5 | $0.00748 | $0.00748 |
Grade A, and why
scopegate CLAUDE.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 — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
@.darkflow.d/claude.md
Commands
pnpm dev # dev server → localhost:3000
pnpm build # prisma generate + migrate deploy + next build (requires live DB)
pnpm test # vitest run
pnpm lint # eslint
bash ~/.darkflow/darkflow-run.sh # Dark Flow routine dispatcher (global worker)
Commits & CI
Do NOT run pnpm build before or after commits. The Dark Flow CI gate (.github/workflows/darkflow-ci-gate.yml) runs install → build (prisma generate && tsc --noEmit) → lint → test on every push/PR on the project's self-hosted runner; it does not file anything itself. The ci-watch routine reads that workflow's conclusion and files a source:ci task on failure. This overrides the global "always run pnpm build before committing" rule because pnpm build runs migrations and requires a live DB. Failing checks → source:ci task → the fix-ci-issue worker pushes a fix (retries up to 3x, then needs-human); the task closes on green. Running pnpm lint/pnpm test locally for fast feedback is fine but not required. See docs/ci-runner.md for the full split between the workflow and ci-watch.
Architecture
Next.js 16 App Router · Prisma 7 + PostgreSQL · Auth: better-auth · MCP server: MCP SDK Streamable HTTP at /api/mcp/[apiKey] · OTel → SigNoz.
src/
app/
(auth)/ # login / register
(dashboard)/ # admin / projects / settings / notifications
api/
mcp/[apiKey]/ # MCP HTTP endpoint (SSE keep-alive pings every 30s)
cron/ # refresh-tokens (CRON_SECRET-guarded)
oauth/ # OAuth callback handlers
admin/ # invites / projects / users management
notifications/ # notification API
csp-report/ # CSP violation sink → SigNoz OTel span + structured log
projects/[projectId]/ # per-project API
lib/
mcp/
tools/ # per-service tool files (gmail, linkedin, stripe, airtable, calendar, calendly, openrouter, telegram, twitter-ads, …)
index.ts # TOOL_DEFINITIONS — aggregates all service tools
types.ts # ToolDefinition · ToolContext
permissions.ts # per-action permission definitions; PERMISSION_GROUPS derived from PROVIDER_REGISTRY
handler.ts # tool execution + audit logging; 30s per-tool timeout; OAuthTokenError detection
metrics.ts # lazy-init OTel counters (mcp.invalid_requests, mcp.blocked_requests)
service-fetch.ts # unified transport — TRANSPORT_CONFIGS derived from PROVIDER_REGISTRY + SSRF-safe fetch
safe-fetch.ts # SSRF/DNS-rebinding-safe fetch (node:https); validates all A/AAAA records
retry.ts # retryWithDelays(), isRetriableNetworkError(), retryAfterDelayMs(), sleep()
rate-limit.ts # checkRateLimit() — atomic upsert into RateLimitBucket; injectable DB for tests
image-utils.ts # image/video download helpers (uses media-body.ts for streaming size limits)
media-body.ts # readBodyWithLimit(res, maxBytes, label) — streaming download with size cap
email-parser.ts # email parsing utilities
gmail.ts # gmailFetch(), listGmailMessages(), getGmailMessage(), attachment helpers
<service>.ts # service-specific helpers (ahrefs, semrush, email, youtube, …)
tools/
fetch-tool.ts # createFetchTool() / createFetchTools() — declarative tool factory for REST endpoints
auth.ts
auth-middleware.ts # AuthError / ForbiddenError / NotFoundError; requireAuth/requireProject helpers
project-auth.ts # project-level auth guards (verifies team membership)
project-roles.ts # PROJECT_ROLE constants; isProjectOwner()
admin.ts # isAdmin() helper (checks ADMIN_EMAIL env var)
audit.ts # recordAudit() — thin wrapper around mcp/audit-utils
crypto.ts # AES-256-GCM encrypt/decrypt (uses BETTER_AUTH_SECRET)
oauth-flow.ts # shared handleOAuthStart/handleOAuthCallback base logic
oauth-state.ts # HMAC-signed OAuth state (uses BETTER_AUTH_SECRET)
oauth-callback-config.ts # OAUTH_CALLBACK_REGISTRY — per-provider config factory (lazy-import)
oauth-callback-route.ts # createOAuthCallbackRoute(routeKey) — collapses OAuth callback boilerplate
oauth-token-lifecycle.ts # unified token refresh for all 26 providers; OAuthTokenError class; derives config from provider-registry
token-refresh.ts # cron token refresh service — PERMANENT_OAUTH_ERRORS, RefreshResult, refreshConnections(); injectable DB/clock for tests
endpoint-permissions.ts # createProjectEndpoint(), applyEndpointPermissions(), deleteEndpoint(), regenerateEndpointKey(), EndpointPermissionError
cron-auth.ts # checkCronAuth(request, secret?) — Bearer-token check for /api/cron/* routes; returns {ok:true} | {ok:false, status:401|500, error}
provider-registry.ts # PROVIDER_REGISTRY — single source of truth for all 26 providers (token strategy, transport, retry, permissions); add/remove a provider here only
db.ts # Prisma client
provider-names.ts # OAuth provider display name map
instrumentation.ts # OTel hook (browser-safe entry point)
instrumentation.node.ts # OTel SDK init + PrismaInstrumentation (Node.js only)
prisma/
schema.prisma # User · Session · Account · Project · TeamMember ·
# ServiceConnection · McpEndpoint · EndpointPermission ·
# AuditLog · RateLimitBucket · Notification · InviteToken
docker/
prisma.config.ts # runtime Prisma config for migrate deploy (supplies DATABASE_URL; no dotenv)
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 · +44 tokens per session d35774e8ab01
- 10d ago First seen · 185 lines · 7,433 tokens per session scan A f4f3e1e118d1
scopegate CLAUDE.md is an instructions file published in the GitHub repository alifanov/scopegate (16 stars, last pushed today), licensed MIT. It adds 7,477 tokens to every session, about $0.0374 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
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.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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).
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
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.
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.