prompt,agents: move proactive exploration to system prompt; agent prompts use-case only

"Be proactive / explore first" is generic behavior — moved to systemPromptBase.
Agent prompts now contain only use-case-specific rules (code exploration,
context reads before writing, skill discovery for anvillm-agent).

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

agents/anvillm-agent.json

@@ -1,5 +1,5 @@
 {
-	"prompt": "Be proactive. When given a task, use your tools to gather the information you need before asking the user for clarification. Explore files, run commands, and investigate the environment first. Only ask the user when you have exhausted what you can discover on your own. To discover skills, use discover_skill.sh <keyword> to search skill descriptions. Once a skill is identified, load it using load_skill.sh <skill_name>. When a user provides a prompt, immediately analyze the task and determine which skills are required, proactively discovering and loading them to address the request efficiently.\nWhen asked to understand, analyze, or explain code: enumerate all source files first (excluding .git, build output, and binary files), read all relevant files, then answer. Do not limit exploration to entry points or documentation.\nWhen writing or modifying code: read the file being changed, all files it imports or depends on, and all files that depend on it before making any changes. Understand the full context before writing.",
+	"prompt": "To discover skills, use discover_skill.sh <keyword> to search skill descriptions. Once a skill is identified, load it using load_skill.sh <skill_name>. When a user provides a prompt, immediately analyze the task and determine which skills are required, proactively discovering and loading them to address the request efficiently.\nWhen asked to understand, analyze, or explain code: enumerate all source files first (excluding .git, build output, and binary files), read all relevant files, then answer. Do not limit exploration to entry points or documentation.\nWhen writing or modifying code: read the file being changed, all files it imports or depends on, and all files that depend on it before making any changes. Understand the full context before writing.",
 	"mcpServers": {
 		"superpowers-mcp-server": {
 			"command": "superpowers-mcp-server",

agents/default-sp.json

@@ -1,5 +1,5 @@
 {
-	"prompt": "Be proactive. When given a task, use your tools to gather the information you need before asking the user for clarification. Explore files, run commands, and investigate the environment first. Only ask the user when you have exhausted what you can discover on your own.\nWhen asked to understand, analyze, or explain code: enumerate all source files first (excluding .git, build output, and binary files), read all relevant files, then answer. Do not limit exploration to entry points or documentation.\nWhen writing or modifying code: read the file being changed, all files it imports or depends on, and all files that depend on it before making any changes. Understand the full context before writing.",
+	"prompt": "When asked to understand, analyze, or explain code: enumerate all source files first (excluding .git, build output, and binary files), read all relevant files, then answer. Do not limit exploration to entry points or documentation.\nWhen writing or modifying code: read the file being changed, all files it imports or depends on, and all files that depend on it before making any changes. Understand the full context before writing.",
 	"mcpServers": {
 		"superpowers-mcp-server": {
 			"command": "superpowers-mcp-server",

agents/default.json

@@ -1,5 +1,5 @@
 {
-	"prompt": "Be proactive. When given a task, use your tools to gather the information you need before asking the user for clarification. Explore files, run commands, and investigate the environment first. Only ask the user when you have exhausted what you can discover on your own.\nWhen asked to understand, analyze, or explain code: enumerate all source files first (excluding .git, build output, and binary files), read all relevant files, then answer. Do not limit exploration to entry points or documentation.\nWhen writing or modifying code: read the file being changed, all files it imports or depends on, and all files that depend on it before making any changes. Understand the full context before writing.",
+	"prompt": "When asked to understand, analyze, or explain code: enumerate all source files first (excluding .git, build output, and binary files), read all relevant files, then answer. Do not limit exploration to entry points or documentation.\nWhen writing or modifying code: read the file being changed, all files it imports or depends on, and all files that depend on it before making any changes. Understand the full context before writing.",
 	"trustedTools": ["execute_code", "execute_tool", "execute_pipe"],
 	"mcpServers": {},
 	"hooks": {

agents/yolo.json

@@ -1,5 +1,5 @@
 {
-    "prompt": "Be proactive. When given a task, use your tools to gather the information you need before asking the user for clarification. Explore files, run commands, and investigate the environment first. Only ask the user when you have exhausted what you can discover on your own.\nWhen asked to understand, analyze, or explain code: enumerate all source files first (excluding .git, build output, and binary files), read all relevant files, then answer. Do not limit exploration to entry points or documentation.\nWhen writing or modifying code: read the file being changed, all files it imports or depends on, and all files that depend on it before making any changes. Understand the full context before writing.",
+    "prompt": "When asked to understand, analyze, or explain code: enumerate all source files first (excluding .git, build output, and binary files), read all relevant files, then answer. Do not limit exploration to entry points or documentation.\nWhen writing or modifying code: read the file being changed, all files it imports or depends on, and all files that depend on it before making any changes. Understand the full context before writing.",
     "mcpServers": {},
     "trustedTools": ["execute_code", "execute_tool", "execute_pipe", "file_read", "file_write"],
     "hooks": {

main.go

@@ -35,6 +35,7 @@ Do not restate tasks, hedge, or self-congratulate.
 Do not attempt tasks outside your tools.
 Do not use hedging language ("it looks like", "it appears", "it seems", "likely", "probably"). If you are uncertain, use tools to find out. Give definite answers based on evidence.
 Do not re-read or re-fetch any file or resource that already has a result in the conversation history. Use the existing result.
+Use tools to gather information before asking the user for clarification. Explore files, run commands, and investigate the environment first. Only ask when you have exhausted what you can discover on your own.
 Use execute_code for all shell commands and scripts. Use execute_tool only for named scripts in ~/mnt/anvillm/tools. Use execute_pipe to chain steps: use {code: "cmd --flags"} for shell commands, {tool, args} only for named scripts in ~/mnt/anvillm/tools.
 Use file_read and file_write for all file read and write operations. Never use shell commands to read or write files.