Add commands

Author: SimonSchubert

assets/commands/ccc.md

@@ -0,0 +1,74 @@
+# TAGLINE
+
+ContextCodeCache generator for AI coding agents
+
+# TLDR
+
+**Scan a project and generate the cache**
+
+```ccc scan```
+
+**Scan and also produce a token stream**
+
+```ccc scan . --tokens```
+
+**Check whether the committed cache is up to date (CI)**
+
+```ccc check .```
+
+**Install the binary to PATH (Linux)**
+
+```ccc install```
+
+# SYNOPSIS
+
+**ccc** scan [_PATH_] [--tokens]
+
+**ccc** check [_PATH_] [--format json]
+
+**ccc** tokenize [_PATH_]
+
+**ccc** install [--dir DIR] [--force]
+
+# DESCRIPTION
+
+ccc scans a source tree and produces a compact, machine-readable `.ccc/` directory containing per-file summaries: constants, functions (with types and doc summaries), call-graph edges inside the file, and marker comments (TODO, FIXME, etc.).
+
+The cache is designed to be committed and consumed by AI agents so they have a cheap, always-fresh structural index instead of repeatedly grepping or re-reading source.
+
+Supported languages: Rust, Python, JavaScript, TypeScript, Go.
+
+# INSTALL
+
+cargo build --release
+./target/release/ccc install
+
+The `ccc install` subcommand copies the binary to `~/.local/bin` (or a directory you choose).
+
+# PARAMETERS
+
+**scan**  
+> Regenerate the `.ccc` directory for the given path (defaults to current directory).
+
+**--tokens**  
+> Also emit a pre-encoded token stream (tokens.bin + tokens.json) for models that can consume raw token IDs.
+
+**check**  
+> Exit non-zero if the on-disk cache is stale relative to current sources. Useful in CI.
+
+**--format json**  
+> Machine-readable output listing changed files.
+
+# CAVEATS
+
+The token stream uses approximate tiktoken IDs and is noted as such; it is not guaranteed to match Anthropic tokenizers exactly. Use the Markdown form of the cache with Claude.
+
+Regenerate the cache whenever source files that affect the index change.
+
+# SEE ALSO
+
+[colwill/ccc](https://github.com/colwill/ccc)
+
+```[Source code](https://github.com/colwill/ccc)
+<!-- verified: 2026-07-04 -->
+```

assets/commands/index.txt

@@ -1052,6 +1052,7 @@ cbt.md
 cc.md
 ccache.md
 ccat.md
+ccc.md
 ccgs.md
 ccomps.md
 ccrypt.md
@@ -4351,6 +4352,7 @@ mcookie.md
 mcopy.md
 mcp-postgres.md
 mcp-probe.md
+mcpsnoop.md
 mcptube.md
 mcs.md
 md5sum.md
@@ -6593,6 +6595,7 @@ ri.md
 rich.md
 riff.md
 rig.md
+rigorix.md
 rip.md
 ripd.md
 ripgrep.md

assets/commands/mcpsnoop.md

@@ -0,0 +1,86 @@
+# TAGLINE
+
+Wireshark for MCP – transparent proxy and live TUI for AI client/server traffic
+
+# TLDR
+
+**Wrap a server** to inspect live MCP (Model Context Protocol) JSON-RPC traffic
+
+```mcpsnoop -- node build/index.js```
+
+**Run the demo**
+
+```mcpsnoop demo```
+
+**Proxy an HTTP MCP server**
+
+```mcpsnoop http --target http://localhost:3000/mcp --listen :7000```
+
+**Start the TUI** (pairs automatically with wrapped servers)
+
+```mcpsnoop```
+
+# SYNOPSIS
+
+**mcpsnoop** [--] [_server_ _args_...]
+
+**mcpsnoop** http [--target _url_] [--listen _addr_]
+
+**mcpsnoop** demo
+
+# DESCRIPTION
+
+mcpsnoop is a transparent proxy and interactive terminal UI for debugging traffic between AI clients (Claude Desktop, Cursor, Claude Code, etc.) and MCP servers. Unlike the official MCP Inspector, which connects as a separate client, mcpsnoop sits in the actual data path so it sees the real calls made by your client.
+
+It consists of two cooperating roles in one binary:
+
+- The shim (`mcpsnoop -- <server command>`) that your client spawns. It forwards bytes verbatim while capturing every JSON-RPC frame.
+- The hub/TUI (`mcpsnoop` with no arguments) that receives frames and provides a live, filterable, replay-capable interface.
+
+The two halves discover each other automatically via a well-known socket and on-disk logs; order of starting does not matter.
+
+# FEATURES
+
+- Live colour-coded JSON-RPC stream (requests, responses, notifications, server stderr)
+- Hung-call detection with live timers
+- Capability inspector
+- Frame inspector with search
+- Replay of any captured tool call against a fresh server copy
+- Powerful filter query language (`tool:`, `status:`, `dir:`, etc.)
+- Single static binary, zero runtime dependencies
+
+# INSTALL
+
+go install github.com/kerlenton/mcpsnoop/cmd/mcpsnoop@latest
+
+Or via Homebrew:
+
+brew tap kerlenton/mcpsnoop
+brew install mcpsnoop
+
+Prebuilt binaries are available from the Releases page.
+
+# PARAMETERS
+
+**--**  
+> End flag processing; everything after is the wrapped server command and its arguments.
+
+For the HTTP reverse-proxy mode:
+
+**--target** _url_  
+> Target HTTP MCP endpoint.
+
+**--listen** _addr_  
+> Address to listen on (e.g. `:7000`).
+
+# CAVEATS
+
+Only wrap servers you trust. mcpsnoop executes the command you configure. The tool is pre-1.0; behaviour may change in minor releases.
+
+# SEE ALSO
+
+[modelcontextprotocol/inspector](https://github.com/modelcontextprotocol/inspector)
+
+```[Source code](https://github.com/kerlenton/mcpsnoop)
+<!-- verified: 2026-07-04 -->
+```

assets/commands/rigorix.md

@@ -0,0 +1,66 @@
+# TAGLINE
+
+Deterministic coding-agent runtime that compiles intent into executable DAGs
+
+# TLDR
+
+**Initialize a project**
+
+```rigorix init```
+
+**Generate and review a plan** before executing
+
+```rigorix plan "Add a new API endpoint"```
+
+**Run a task**
+
+```rigorix run "Refactor the user module"```
+
+**Show the dashboard / interactive TUI**
+
+```rigorix```
+
+# SYNOPSIS
+
+**rigorix** [_options_] <command> [_args_...]
+
+Common commands: init, plan, run, update
+
+# DESCRIPTION
+
+Rigorix turns natural-language development tasks into deterministic, auditable Directed Acyclic Graphs (DAGs). Planning is separated from execution: an LLM generates a plan according to templates; Rigorix validates the plan, then executes it under policy, permission, and budget controls.
+
+Key benefits over open-ended agent loops:
+
+- Repeatable execution graphs
+- Explicit quality gates and policy enforcement
+- Full audit trail
+- Safe to run in CI/CD contexts
+
+It supports Rust, TypeScript, Python and Go codebases (TypeScript is the most mature).
+
+# INSTALL
+
+cargo install --git https://github.com/arman-jalili/rigorix-oss rigorix-cli
+
+Or build from source and place the rigorix binary on your PATH.
+
+Set your LLM API key, e.g.:
+
+export RIGORIX__LLM__API_KEY="sk-ant-..."
+
+# PARAMETERS
+
+Most configuration lives in the project after `rigorix init`. Command line flags control planning vs. execution, provider selection, and non-interactive mode.
+
+# CAVEATS
+
+Rigorix is intentionally less flexible than free-form chat agents. It excels at repeatable, governed workflows rather than exploratory conversation.
+
+# SEE ALSO
+
+[arman-jalili/rigorix-oss](https://github.com/arman-jalili/rigorix-oss)
+
+```[Source code](https://github.com/arman-jalili/rigorix-oss)
+<!-- verified: 2026-07-04 -->
+```