@mcp-s/cli 0.0.11 → 0.0.13

package/README.md

@@ -76,7 +76,7 @@ mcp-s-cli call <tool_name> '{"param": "value"}'
 mcp-s-cli [options]                        List all tools
 mcp-s-cli [options] info                   Show server details
 mcp-s-cli [options] info <tool>            Show schema for a specific tool
-mcp-s-cli [options] grep <pattern>         Search tools by glob pattern
+mcp-s-cli [options] grep <query>           Search tools by name
 mcp-s-cli [options] call <tool>            Call a tool (reads JSON args from stdin)
 mcp-s-cli [options] call <tool> <json>     Call a tool with inline JSON arguments
 mcp-s-cli init [--org <org>] [--base-url <url>] [...]  Initialize global config
@@ -130,13 +130,13 @@ server
 ### Search Tools
 
 ```bash
-# Find tools matching a glob pattern
-$ mcp-s-cli grep "*ticket*"
+# Find tools by name (case-insensitive substring match)
+$ mcp-s-cli grep ticket
 create_ticket
 update_ticket
 
 # With descriptions
-$ mcp-s-cli grep "*search*" -d
+$ mcp-s-cli grep search -d
 search_issues - Search Jira issues by query
 ```
 
@@ -406,7 +406,7 @@ Commands:
   mcp-s-cli                    # List all tools
   mcp-s-cli info               # Show server details
   mcp-s-cli info <tool>        # Get tool schema
-  mcp-s-cli grep "<pattern>"   # Search tools by name
+  mcp-s-cli grep <query>       # Search tools by name
   mcp-s-cli call <tool>        # Call tool (stdin for JSON args)
   mcp-s-cli call <tool> '{}'   # Call tool with inline JSON
 
@@ -446,7 +446,7 @@ Each CLI invocation connects to the single configured server.
 | Command                      | Behaviour                         |
 | ---------------------------- | --------------------------------- |
 | `mcp-s-cli` (list all)       | Connects to the configured server |
-| `mcp-s-cli grep "<pattern>"` | Connects to the configured server |
+| `mcp-s-cli grep <query>`     | Connects to the configured server |
 | `mcp-s-cli info`             | Connects to the configured server |
 | `mcp-s-cli info <tool>`      | Connects to the configured server |
 | `mcp-s-cli call <tool> '{}'` | Connects to the configured server |

package/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: mcp-s-cli
-description: Interface for MCP (Model Context Protocol) servers via CLI. Use when you need to interact with external tools, APIs, or data sources through MCP servers.
+description: Use when you need to interact with external tools, APIs, or data sources through MCP servers.
 ---
 
 # mcp-s-cli
@@ -9,17 +9,17 @@ Access a single MCP server through the command line. MCP enables interaction wit
 
 ## Commands
 
-| Command                          | Output                                              |
-| -------------------------------- | --------------------------------------------------- |
-| `mcp-s-cli`                      | List all available tools                            |
-| `mcp-s-cli info <tool>`          | Get tool JSON schema                                |
-| `mcp-s-cli grep "<pattern>"`     | Search tools by name (names only, not descriptions) |
-| `mcp-s-cli call <tool>`          | Call tool (reads JSON from stdin if no args)        |
-| `mcp-s-cli call <tool> '<json>'` | Call tool with arguments                            |
+| Command                          | Output                                            |
+| -------------------------------- | ------------------------------------------------- |
+| `mcp-s-cli`                      | List all available tools                          |
+| `mcp-s-cli info <tool>`          | Get tool JSON schema                              |
+| `mcp-s-cli grep <query>`         | Search tools by name (case-insensitive substring) |
+| `mcp-s-cli call <tool>`          | Call tool (reads JSON from stdin if no args)      |
+| `mcp-s-cli call <tool> '<json>'` | Call tool with arguments                          |
 
 ## Workflow
 
-1. **Discover**: `mcp-s-cli grep "<pattern>"` → find tools by keyword; fall back to `mcp-s-cli` only when you need a full inventory
+1. **Discover**: `mcp-s-cli grep <query>` → find tools by keyword; fall back to `mcp-s-cli` only when you need a full inventory
 2. **Inspect**: `mcp-s-cli info <tool>` → get full JSON schema
 3. **Execute**: `mcp-s-cli call <tool> '<json>'` → run with arguments
 
@@ -29,7 +29,7 @@ Every command consumes tokens. Pick the cheapest operation that gets the job don
 
 ### Discovery: start narrow
 
-- **`grep "<pattern>"`** — cheapest. Use when you can guess a keyword from the user's request (e.g., "read a github file" → `grep "file"` or `grep "content"`).
+- **`grep <query>`** — cheapest. Use when you can guess a keyword from the user's request (e.g., "read a github file" → `grep file` or `grep content`).
 - **`mcp-s-cli`** (names only) — moderate. Use when you need a full inventory (e.g., "what tools do I have?" or "analyze logs from all tools").
 - **`mcp-s-cli -d`** (names + descriptions) — most expensive. Use only when names alone aren't enough to pick the right tool.
 
@@ -41,11 +41,21 @@ Every command consumes tokens. Pick the cheapest operation that gets the job don
 - **Pipe output** through shell tools (`head`, `grep`, `jq`) to trim large responses before they enter context.
 - Never fetch a full list when the user only needs a few fields (names, links, IDs).
 
+### When a filter produces no output
+
+When you pipe through a filter and get empty output, **do not** fall back to the raw unfiltered call. Use the **probe-then-filter** pattern:
+
+1. **Probe**: fetch a _single_ item (e.g., `limit: 1`, or a dedicated get-one tool) without any pipe.
+2. **Learn**: read the response to understand the actual field names, nesting, and types.
+3. **Re-run**: call the bulk tool again with a corrected filter based on what you learned.
+
 ### Decision heuristic
 
 - User mentions a specific tool or domain keyword → `grep` for it.
 - User asks a broad question across all tools → list tools first, then act.
+- Task involves multiple items of the same type → use one bulk/list call, not N individual calls.
 - Tool may return large payloads → use filter params + pipe to trim output.
+- Filter script produced no output → **probe one item** to learn the shape, then re-run with a corrected filter. Never fall back to the raw unfiltered call.
 
 ## Examples
 
@@ -66,7 +76,7 @@ mcp-s-cli call my-tool '{"arg-name": "arg-value"}'
 cat args.json | mcp-s-cli call my-tool
 
 # Search for tools
-mcp-s-cli grep "*file*"
+mcp-s-cli grep file
 
 # Output is raw text (pipe-friendly)
 mcp-s-cli call read_file '{"path": "./file"}' | head -10

package/dist/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: mcp-s-cli
-description: Interface for MCP (Model Context Protocol) servers via CLI. Use when you need to interact with external tools, APIs, or data sources through MCP servers.
+description: Use when you need to interact with external tools, APIs, or data sources through MCP servers.
 ---
 
 # mcp-s-cli
@@ -9,17 +9,17 @@ Access a single MCP server through the command line. MCP enables interaction wit
 
 ## Commands
 
-| Command                          | Output                                              |
-| -------------------------------- | --------------------------------------------------- |
-| `mcp-s-cli`                      | List all available tools                            |
-| `mcp-s-cli info <tool>`          | Get tool JSON schema                                |
-| `mcp-s-cli grep "<pattern>"`     | Search tools by name (names only, not descriptions) |
-| `mcp-s-cli call <tool>`          | Call tool (reads JSON from stdin if no args)        |
-| `mcp-s-cli call <tool> '<json>'` | Call tool with arguments                            |
+| Command                          | Output                                            |
+| -------------------------------- | ------------------------------------------------- |
+| `mcp-s-cli`                      | List all available tools                          |
+| `mcp-s-cli info <tool>`          | Get tool JSON schema                              |
+| `mcp-s-cli grep <query>`         | Search tools by name (case-insensitive substring) |
+| `mcp-s-cli call <tool>`          | Call tool (reads JSON from stdin if no args)      |
+| `mcp-s-cli call <tool> '<json>'` | Call tool with arguments                          |
 
 ## Workflow
 
-1. **Discover**: `mcp-s-cli grep "<pattern>"` → find tools by keyword; fall back to `mcp-s-cli` only when you need a full inventory
+1. **Discover**: `mcp-s-cli grep <query>` → find tools by keyword; fall back to `mcp-s-cli` only when you need a full inventory
 2. **Inspect**: `mcp-s-cli info <tool>` → get full JSON schema
 3. **Execute**: `mcp-s-cli call <tool> '<json>'` → run with arguments
 
@@ -29,7 +29,7 @@ Every command consumes tokens. Pick the cheapest operation that gets the job don
 
 ### Discovery: start narrow
 
-- **`grep "<pattern>"`** — cheapest. Use when you can guess a keyword from the user's request (e.g., "read a github file" → `grep "file"` or `grep "content"`).
+- **`grep <query>`** — cheapest. Use when you can guess a keyword from the user's request (e.g., "read a github file" → `grep file` or `grep content`).
 - **`mcp-s-cli`** (names only) — moderate. Use when you need a full inventory (e.g., "what tools do I have?" or "analyze logs from all tools").
 - **`mcp-s-cli -d`** (names + descriptions) — most expensive. Use only when names alone aren't enough to pick the right tool.
 
@@ -41,11 +41,21 @@ Every command consumes tokens. Pick the cheapest operation that gets the job don
 - **Pipe output** through shell tools (`head`, `grep`, `jq`) to trim large responses before they enter context.
 - Never fetch a full list when the user only needs a few fields (names, links, IDs).
 
+### When a filter produces no output
+
+When you pipe through a filter and get empty output, **do not** fall back to the raw unfiltered call. Use the **probe-then-filter** pattern:
+
+1. **Probe**: fetch a _single_ item (e.g., `limit: 1`, or a dedicated get-one tool) without any pipe.
+2. **Learn**: read the response to understand the actual field names, nesting, and types.
+3. **Re-run**: call the bulk tool again with a corrected filter based on what you learned.
+
 ### Decision heuristic
 
 - User mentions a specific tool or domain keyword → `grep` for it.
 - User asks a broad question across all tools → list tools first, then act.
+- Task involves multiple items of the same type → use one bulk/list call, not N individual calls.
 - Tool may return large payloads → use filter params + pipe to trim output.
+- Filter script produced no output → **probe one item** to learn the shape, then re-run with a corrected filter. Never fall back to the raw unfiltered call.
 
 ## Examples
 
@@ -66,7 +76,7 @@ mcp-s-cli call my-tool '{"arg-name": "arg-value"}'
 cat args.json | mcp-s-cli call my-tool
 
 # Search for tools
-mcp-s-cli grep "*file*"
+mcp-s-cli grep file
 
 # Output is raw text (pipe-friendly)
 mcp-s-cli call read_file '{"path": "./file"}' | head -10

package/dist/daemon.js

@@ -414,7 +414,7 @@ import { join as join2 } from "path";
 import { fileURLToPath } from "url";
 
 // src/version.ts
-var VERSION = "0.0.11";
+var VERSION = "0.0.13";
 
 // src/client.ts
 function getRetryConfig(settings) {

package/dist/index.js

@@ -1298,7 +1298,7 @@ async function cleanupOrphanedDaemons() {
 }
 
 // src/version.ts
-var VERSION = "0.0.11";
+var VERSION = "0.0.13";
 
 // src/client.ts
 function getRetryConfig(settings) {
@@ -2183,32 +2183,8 @@ function configCommand(opts) {
 }
 
 // src/commands/grep.ts
-function globToRegex(pattern) {
-  let escaped = "";
-  let i = 0;
-  while (i < pattern.length) {
-    const char = pattern[i];
-    if (char === "*" && pattern[i + 1] === "*") {
-      escaped += ".*";
-      i += 2;
-      while (pattern[i] === "*") {
-        i++;
-      }
-    } else if (char === "*") {
-      escaped += "[^/]*";
-      i += 1;
-    } else if (char === "?") {
-      escaped += "[^/]";
-      i += 1;
-    } else if ("[.+^${}()|\\]".includes(char)) {
-      escaped += `\\${char}`;
-      i += 1;
-    } else {
-      escaped += char;
-      i += 1;
-    }
-  }
-  return new RegExp(`^${escaped}$`, "i");
+function matchesToolName(name, query) {
+  return name.toLowerCase().includes(query.toLowerCase());
 }
 async function grepCommand(options) {
   let loaded;
@@ -2220,8 +2196,7 @@ async function grepCommand(options) {
   }
   const { serverConfig, settings } = loaded;
   const serverLabel = "server";
-  const pattern = globToRegex(options.pattern);
-  debug(`Searching for pattern "${options.pattern}"`);
+  debug(`Searching for "${options.pattern}"`);
   let connection = null;
   const allResults = [];
   let searchError;
@@ -2229,7 +2204,7 @@ async function grepCommand(options) {
     connection = await getConnection(serverLabel, serverConfig, settings);
     const tools = await connection.listTools();
     for (const tool of tools) {
-      if (pattern.test(tool.name)) {
+      if (matchesToolName(tool.name, options.pattern)) {
         allResults.push({ server: serverLabel, tool });
       }
     }
@@ -2247,8 +2222,9 @@ async function grepCommand(options) {
   }
   if (allResults.length === 0) {
     console.log(`No tools found matching "${options.pattern}"`);
-    console.log("  Tip: Pattern matches tool names only");
-    console.log(`  Tip: Use '*' for wildcards, e.g. '*file*' or 'read_*'`);
+    console.log(
+      "  Tip: Search matches tool names only (substring, case-insensitive)"
+    );
     console.log(`  Tip: Run 'mcp-s-cli' to list all available tools`);
     return;
   }
@@ -3313,7 +3289,7 @@ Usage:
   mcp-s-cli [options]                        List all tools
   mcp-s-cli [options] info                   Show server details
   mcp-s-cli [options] info <tool>            Show tool schema
-  mcp-s-cli [options] grep <pattern>         Search tools by glob pattern
+  mcp-s-cli [options] grep <query>           Search tools by name
   mcp-s-cli [options] call <tool>            Call tool (reads JSON from stdin if no args)
   mcp-s-cli [options] call <tool> <json>     Call tool with JSON arguments
   mcp-s-cli init --org <org> [--mcp <id>] [--toolkit <tk>] [--user-access-key <key>]  Initialize with org
@@ -3344,14 +3320,14 @@ Options:
       --user-access-key <key>  User access key \u2014 triggers stdio mode (used with init)
 
 Output:
-  mcp-s-cli/info/grep        Human-readable text to stdout
+  mcp-s-cli/info/grep       Human-readable text to stdout
   call                     Raw JSON to stdout (for piping)
   Errors                   Always to stderr
 
 Examples:
   mcp-s-cli                                  # List all tools
   mcp-s-cli -d                               # List with descriptions
-  mcp-s-cli grep "*file*"                    # Search for file tools
+  mcp-s-cli grep file                        # Search for file tools
   mcp-s-cli info                             # Show server details
   mcp-s-cli info read_file                   # Show tool schema
   mcp-s-cli call read_file '{}'              # Call tool

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@mcp-s/cli",
-  "version": "0.0.11",
+  "version": "0.0.13",
   "description": "A lightweight CLI for connecting AI agents to the Webrix MCP Gateway",
   "type": "module",
   "main": "dist/index.js",