DeskcommCRM
melgarafael/DeskcommCRM/AGENTS.md
clínicas, imobiliárias, infoprodutos, serviços), WhatsApp como canal primário via WAHA, CRM inteiro exposto por MCP. Multi-tenant com RLS desde o dia 1; LGPD nativa. Monetização = **self-host
How real projects brief Codex, Cursor and every other agent that reads AGENTS.md.
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
melgarafael/DeskcommCRM/AGENTS.md
clínicas, imobiliárias, infoprodutos, serviços), WhatsApp como canal primário via WAHA, CRM inteiro exposto por MCP. Multi-tenant com RLS desde o dia 1; LGPD nativa. Monetização = **self-host
mock-server/mockserver/AGENTS.md
response builder, provider codecs, streaming physics, conversation matchers, isolation, MCP tools, or LLM dashboard | | [docs/code/ai-protocol-mocking.md](docs/code/ai-protocol-mocking.md) | Before modifying MCP/A2A server mocking, SSE/WebSocket/JSON-RPC mock responses, or AI protocol handlers | | [docs/code/llm-codec-fixtures.md](docs/code/llm-codec-fixtures.md
aldinokemal/go-whatsapp-web-multidevice/AGENTS.md
WhatsApp API server using whatsmeow, Fiber, and SQLite. The Go module and local runtime directory are `src/`. `rest` also mounts MCP at `/mcp` when enabled; both transports share the same
tfutils/tfenv/AGENTS.md
repos/tfutils/tfenv/...` for operations not covered by high-level `gh` commands. The GitHub MCP server may supplement `gh` where proven reliable, but must never be on the critical path. --- ## Work Type
campfirein/byterover-cli/AGENTS.md
campfirein/brv-transport-client` - Agent pool manages forked child processes per project; task routing in `server/infra/process/` - MCP server in `server/infra/mcp/` exposes tools via Model Context Protocol; `tools/` subdir has dedicated implementations (`brv-query
withcoral/coral/AGENTS.md
entity ID is globally unique. The one exception is Agent authentication at the Organization MCP Endpoint: it finds the Registration by credential verifier alone, because the credential is a secret
anonaddy/anonaddy/AGENTS.md
explicitly requested by the user. === boost rules === # Laravel Boost ## Tools - Laravel Boost is an MCP server with tools designed specifically for this application. Prefer Boost tools over manual alternatives like
benjitaylor/agentation/AGENTS.md
Monorepo containing: 1. **npm package** (`package/`) - See `package/AGENTS.md` 2. **Website/docs** (`package/example/`) - See `package/example/AGENTS.md` 3. **MCP server** (`mcp/`, published as `agentation-mcp`) 4. **Agent skills** (`skills/`) ## What is Agentation? A floating
mvanhorn/cli-printing-press/AGENTS.md
runtime walker in `internal/mcp/cobratree/` mirrors the Cobra tree at server start and registers every user-facing command as an MCP tool unless one of these applies: 1. Commands annotated `cmd.Annotations
dmno-dev/varlock/AGENTS.md
built from `packages/encryption-binary-swift` (darwin) and `packages/encryption-binary-rust` (linux/win32) - `packages/utils`, `packages/plugins` — shared internals - `packages/varlock-docs-mcp` — docs MCP server for external varlock users; do **not** use it to look things up while working
Steady19 mentions in 2 weeks
No other project uses it yet
eastlondoner/vibe-tools/AGENTS.md
task_id"`) **Model Context Protocol (MCP) Commands:** Use the following commands to interact with MCP servers and their specialized tools: `vibe-tools mcp search " "` - Search the MCP Marketplace for available
usesend/useSend/AGENTS.md
connected to this repository and available to coding agents through the [Polylane MCP server](https://mcp.polylane.com/mcp). - When a question involves production behaviour (an error, a spike, a deploy
sebastienros/jint/Jint.Browser.Mcp/AGENTS.md
Agent instructions: the Model Context Protocol server > **Read this when:** You are touching anything under `Jint.Browser.Mcp/`, the `mcp` command in > `Jint.Browser.Tool/`, or its tests in `Jint.Tests.Browser/Mcp/`. > > This
Steady8 mentions in 2 weeks
No other project uses it yet
sebastienros/jint/Jint.Browser.Tool/AGENTS.md
that command prints no banner and every diagnostic goes to standard error; and `mcp` **hardens the pages by default**, which is the opposite of every other command here
latitude-dev/latitude-llm/AGENTS.md
agents/skills/fix-datadog-issues/SKILL.md](.agents/skills/fix-datadog-issues/SKILL.md) | Finding, triaging, and fixing production errors from **Datadog Error Tracking** (`plugin:datadog:mcp`); picking which issue to work (occurrence/trend/recency, v2-only, prod-only), root-causing in code, reproducing
IBM/mcp-context-forge/AGENTS.md
Server and **never forwarded upstream** — only the exchanged token is sent to the downstream MCP server. - `token_url` on a `token-exchange` gateway is an SSRF / egress boundary: the user
IBM/mcp-context-forge/plugins/AGENTS.md
free once that release lands. ## External Plugins (MCP) External plugins run as separate MCP servers. ```yaml plugins: - name: "ExternalFilter" kind: "external" priority: 10 mcp: proto: STREAMABLEHTTP # or STDIO
IBM/mcp-context-forge/tests/AGENTS.md
current stack (requires test-specific plugin config) make test-mcp-session-isolation-load # Locust MCP session isolation correctness load test # Selective runs pytest -k "fragment" # By name substring pytest
dimensionalOS/dimos/AGENTS.md
same, backgrounded dimos --replay run unitree-go2-agentic # + LLM agent (gpt-5.6-luna) + skills + MCP server dimos run unitree-go2-agentic --robot-ip 192.168.123.161 # real Go2 hardware # --- G1 humanoid --- dimos
kirodotdev/KiroCrew/AGENTS.md
context-management](docs/architecture/context-management.md) + [memory-skills-hooks: Context Builder](docs/system-specs/modules/memory-skills-hooks.md#context-builder-contextpy) (module ownership) | | MCP servers or tools (adding, changing, statelessness) | [mcp](docs/architecture/mcp.md) | | `mcp_gateway/` — `gatewayd.py`, `daemon/`, the daemon
An open format for instructions to coding agents, read by Codex, Cursor and others. Think of it as a README written for agents.
At the repository root, with more specific files in subdirectories. Agents read the one closest to the file they're editing.
Setup and test commands, code style, and the rules a new contributor would need to know.
Claude Code reads CLAUDE.md. A one-line CLAUDE.md that points at AGENTS.md covers both.