obsidian-mind
breferrari/obsidian-mind/AGENTS.md
vsearch|get|multi-get`** — CLI fallback for one-off shell checks or when the MCP server is unavailable. Always pass `--index `, where ` ` is the `qmd_index` field from `vault-manifest.json` when
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.
breferrari/obsidian-mind/AGENTS.md
vsearch|get|multi-get`** — CLI fallback for one-off shell checks or when the MCP server is unavailable. Always pass `--index `, where ` ` is the `qmd_index` field from `vault-manifest.json` when
Marker-Inc-Korea/AutoRAG/AGENTS.md
discrawl, msgvault, …) in place. No forced ingestion into a central index, no third-party server holding a copy of the corpus. Results carry source-native identities (`/kakao/ /chunks/ `), and secrets
debba/tabularis/AGENTS.md
only module behind the `desktop` cargo feature; `cargo build --no-default-features` builds the server without Tauri. - **A desktop command is a thin wrapper over `application/`.** Implement the behaviour there
metalbear-co/mirrord/AGENTS.md
launches the user's local process with the mirrord layer loaded. - `mirrord-mcp`: the MCP server behind `mirrord mcp`, which exposes mirrord to AI agents over stdio. ## Command Reference ### Compiling
hahwul/dalfox/AGENTS.md
/only-discovery/result rendering), `poc.rs`, `postprocess.rs`, `logging.rs` - `src/cmd/mod.rs`: shared `error_codes` constants (used by CLI + server + MCP) - `src/config.rs`: config schema + precedence (`apply_to_scan_args_if_default`) - `src/scanning/`: reflection/DOM checks, AST integration
hahwul/dalfox/docs/AGENTS.md
invariants in the repo-root `AGENTS.md`. Treat it as a fourth interface alongside CLI / server / MCP. - **Output formats**: `src/scanning/result/` and `src/cmd/scan/output.rs` (output routing). Keep `docs/content/guide/output.md` aligned. - **Stage names**: see `src/lib.rs
mistralai/mistral-vibe/AGENTS.md
Agent Client Protocol; `vibe/setup` holds auth and onboarding config shared by the app server and `vibe-acp`. Tests live in `tests/` with autouse fixtures in `conftest.py` and test doubles
Steady17 mentions in 2 weeks
No other project uses it yet
modelcontextprotocol/go-sdk/AGENTS.md
Packages - `mcp`: The core package defining the primary APIs for constructing and using MCP clients and servers. This is where most logic resides. - `jsonrpc`: Provides the JSON-RPC 2.0 transport
No other project uses it yet
homarr-labs/homarr/AGENTS.md
maintainer explicitly runs through `homarr-triage` keep that disclaimer as their first line. ## MCP servers The [Homarr Mantine plugin](plugins/homarr-mantine/plugin.json) bundles the UI skill and pinned documentation MCP server. `.agents/plugins/marketplace.json
No other project uses it yet
pollinations/pollinations/AGENTS.md
React frontend - `packages/sdk/` — `@pollinations/sdk` (client + React hooks) - `packages/mcp/` — shared handlers for the hosted stateless MCP server (see `packages/mcp/AGENTS.md`) - `shared/` — auth, registry, IP queue; `shared/registry/` holds model registries - `apps/` — Applications maintained
Steady3 mentions in 2 weeks
No other project uses it yet
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
Steady6 mentions in 2 weeks
No other project uses it yet
chakra-ui/zag/AGENTS.md
pnpm test` | | Start React dev server | `pnpm start-react` (port 3000) | | Start Preact dev server | `pnpm start-preact` (port 5173) | | E2E tests (React) | `pnpm e2e-react` | | E2E tests (Preact) | `pnpm
apache/iggy/AGENTS.md
types (IggyDuration, IggyTimestamp, ...) │ ├── binary_protocol/ Wire format (stable) │ ├── configs/ Config plumbing + ConfigEnv derive │ ├── ai/mcp/ MCP server │ ├── bench/ Benchmark suite │ ├── integration/ Cross-crate integration tests │ ├── simulator/ Deterministic simulator │ ├── consensus/ Raft (Miri
gdamore/tcell/AGENTS.md
gopls` MCP server by default when working on Go code in this repository, unless I explicitly
icip-cas/PPTAgent/AGENTS.md
generation and evaluation library, still shipped in the package and still used for the MCP server entrypoint `pptagent-mcp`. When making changes, treat `deeppresenter` as the primary product surface unless
IsmaelMartinez/teams-for-linux/AGENTS.md
Portfolio dashboard:** https://ismaelmartinez.github.io/repo-butler/ **Consumer guide:** https://github.com/IsmaelMartinez/repo-butler/blob/main/docs/consumer-guide.md ### Querying Reginald (the butler MCP server) To query your repo's health tier, governance findings, and portfolio data from
transformerlab/transformerlab-app/AGENTS.md
local development: ```bash cd lab-sdk && pip install -e . ``` Then restart the API server. This is a common gotcha — the API imports the *installed* `lab` package, not the source tree
totec448-spec/chat-on-steroids/AGENTS.md
browser page | MCP via public tunnel | native UI and conversation state v | MAIN: fiber.js + usage.js Core / Desktop / Plugins | isolated: chatgpt-dom.js + content.js | secret path per surface v server -> registrar -> kernel background.js
facebook/memlab/AGENTS.md
change it (write a leak test scenario, analyze a heap snapshot, set up the MCP server), read [AI.md](./AI.md) and the Agent Skills under [`plugins/memlab/skills/`](./plugins/memlab/skills) instead. ## Repository layout memlab
clidey/whodb/AGENTS.md
browser AI agent feature (under `ee/`), NOT the coding-agent (Claude/Codex) configuration or any MCP server. When in doubt, ask which is meant before acting. ## Non-Negotiable Rules 1. **GraphQL
No other project uses it yet
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.