CLAUDE.md examples from real projects

SadeghHayeriDeletes or force-pushesCLAUDE.md

GreenTunnel

SadeghHayeri/GreenTunnel/CLAUDE.md

service configured with just one of the two. Entries now carry a separate `secure` endpoint. 6. **No crash-safety whatsoever.** `SIGKILL`, a force-quit or a `npm run dev` reload

pdfmeDeletes or force-pushesCLAUDE.md

pdfme

pdfme/pdfme/CLAUDE.md

only display - `form`: Interactive input - `designer`: Template creation Base class: `packages/ui/src/class.ts` #### 5. Expression System Secure JavaScript expression evaluator for dynamic content: - Uses Acorn for parsing - AST validation for security - Cached

No other project uses it yet

rivet-devCLAUDE.md

agent-os

rivet-dev/agent-os/CLAUDE.md

enter default builds, CI, publication, or behavioral-parity requirements without a separately approved design. ## Security Model Trust model: - **Client**: trusted, except for code/payloads it submits for execution. - **Sidecar/runtime**: trusted enforcement

rivet-devReads credentialsCLAUDE.md

agent-os / crates

rivet-dev/agent-os/crates/CLAUDE.md

architectural invariant. Any path where guest code reaches the real host is a critical security violation. **NEVER use `Command::new("node")` for guest code execution.** Guest JavaScript runs inside

ruvnetReads credentialsCLAUDE.md

ruvector

ruvnet/ruvector/CLAUDE.md

lint ``` - ALWAYS run tests after making code changes - ALWAYS verify build succeeds before committing ## Security Rules - NEVER hardcode API keys, secrets, or credentials in source files - NEVER commit .env files

JetBrainsCLAUDE.md

koog / a2a

JetBrains/koog/a2a/CLAUDE.md

task updates - **Multiple Transport Protocols**: JSON-RPC, HTTP+JSON/REST with extensibility for gRPC - **Authentication & Security**: OpenAPI 3.0 compatible security schemes ### Submodules Architecture 1. **a2a-core**: Core abstractions, data models

No other project uses it yet

prestDeletes or force-pushesCLAUDE.md

prest

prest/prest/.claude/CLAUDE.md

opus"` only for architecture, new dialects, cross-cutting refactors, unknown-root-cause debugging or security-sensitive changes. Escalate only with a stated reason. **Reference:** `.cursor/rules/planning-and-delegation.mdc` ## When to Apply Rules

No other project uses it yet

SupabaseDeletes or force-pushesCLAUDE.md

supabase-js

supabase/supabase-js/.claude/CLAUDE.md

cutting migration notes (one H2 per theme). Per-package migrations live at `packages/core/ /migrations/`. > - **[Security Policy](https://github.com/supabase/supabase-js/security/policy)** - Security policies and responsible disclosure ## Repository Architecture ### Monorepo Structure ```text supabase

taikoxyzCLAUDE.md

taiko-mono

taikoxyz/taiko-mono/CLAUDE.md

DevTools and SvelteKit's built-in debugging - **General**: Use proper debuggers over console.log debugging ## 🔒 Security Best Practices - Never commit sensitive data (private keys, API keys, etc.) - Always validate user inputs

kstenerudPipes a download into a shellCLAUDE.md

KSCrash / Benchmarks

kstenerud/KSCrash/Benchmarks/CLAUDE.md

mise install` (installs Tuist version from config) - **Trust config**: `mise trust` (required once for security) - **Run Tuist**: `mise exec -- tuist ` or activate mise in your shell ## Benchmarks Workflow ### Quick Start

kstenerudPipes a download into a shellCLAUDE.md

KSCrash / Samples

kstenerud/KSCrash/Samples/CLAUDE.md

mise install` (installs Tuist version from config) - **Trust config**: `mise trust` (required once for security) - **Run Tuist**: `mise exec -- tuist ` or activate mise in your shell ## Sample App Workflow

bjarneoPipes a download into a shellCLAUDE.md

cliamp

bjarneo/cliamp/CLAUDE.md

make coverage` | Tests with a coverage profile and a summary for each function | | `make security` | `govulncheck ./...` | | `make ci` | What CI runs: `fmt-check`, `tidy-check`, `vet`, `staticcheck`, `security`, race tests

archtechxReads credentialsCLAUDE.md

tenancy

archtechx/tenancy/CLAUDE.md

databases and users - **TenantDatabaseManagers** - Database-specific implementations (MySQL, PostgreSQL, SQLite, SQL Server) - **Row Level Security (RLS)** - PostgreSQL-based tenant isolation using policies **Advanced Features:** - **Resource Syncing** - Sync central models

vijaythecoderCLAUDE.md

awesome-claude-agents

vijaythecoder/awesome-claude-agents/CLAUDE.md

clarification ### Tool Inheritance - Omitting `tools` field = inherit all tools (recommended) - Specify tools only for security restrictions - Includes WebFetch, MCP tools when available ## Development Guidelines 1. **Creating New Agents**: - Use templates/agent-template.md

yuzutechCLAUDE.md

kroki

yuzutech/kroki/CLAUDE.md

dependency updates that are user-visible (diagram library versions, base Docker image versions) or security-relevant. - Run the linter (`npm run lint` in `mermaid/`, i.e. `biome check src`) whenever

nyldnReads credentialsCLAUDE.md

claude-octopus

nyldn/claude-octopus/CLAUDE.md

mode ✅ Deliver Phase: Reviewing authentication implementation Providers: 🔴 Codex CLI - Code quality analysis 🧭 Antigravity CLI - Security and edge cases 🔵 Claude - Synthesis and recommendations ``` **Debate:** ``` 🐙 **CLAUDE OCTOPUS ACTIVATED** - AI Debate Hub 🐙 Debate

No other project uses it yet

eunomia-bpfCLAUDE.md

bpf-developer-tutorial

eunomia-bpf/bpf-developer-tutorial/CLAUDE.md

Advanced libbpf development - `src/19-21,29,41-42`: Networking (LSM, TC, XDP, sockops) - `src/22-28,34`: Security topics - `src/31,37`: Language integration (Go, Rust) - `src/44-45`: BPF schedulers - `src/47`: GPU tracing - Each tutorial

tigerless-labsDeletes or force-pushesCLAUDE.md

agent-memory

tigerless-labs/agent-memory/CLAUDE.md

behavior primes it. Reserve explicit prohibitions for failure modes actually observed in traces. ## Security — agents run autonomously, so the system must fail safe - **Trust nothing by default.** Treat external input

txn2CLAUDE.md

kubefwd

txn2/kubefwd/CLAUDE.md

injecting the OS call (see `runCommand` and `goos` in `pkg/fwdapi/network.go`), not left uncovered. - **Security**: `make security` runs gosec, govulncheck on both modules (gated by `scripts/govulncheck-gate.py` against `.govulncheck-allow.txt`), and osv-scanner

No other project uses it yet

langroidCLAUDE.md

langroid

langroid/langroid/CLAUDE.md

transforming and passing messages - **Collaboration**: Multiple agents can work together on complex tasks ### Key Security Features: - The `full_eval` flag in both `TableChatAgentConfig` and `VectorStoreConfig` controls code injection protection - Defaults

No other project uses it yet