diff --git a/README.md b/README.md
index 1bf342c..cfac78e 100644
--- a/README.md
+++ b/README.md
@@ -80,7 +80,17 @@ sudo chmod +x /usr/local/bin/coding-context
 
 ```
 Usage:
-  coding-context [options] <task-name>
+  coding-context [options] <task-name> [user-prompt]
+
+Arguments:
+  <task-name>
+      The name of a task file to look up in task search paths (.agents/tasks).
+      Task files are matched by filename (without .md extension).
+  
+  [user-prompt] (optional)
+      Optional text to append to the task. It can contain slash commands
+      (e.g., '/command-name') which will be expanded, and parameter 
+      substitution (${param}).
 
 Options:
   -C string
@@ -96,7 +106,7 @@ Options:
     	Include rules with matching frontmatter. Can be specified multiple times as key=value.
     	Note: Only matches top-level YAML fields in frontmatter.
   -a string
-    	Default agent to use if task doesn't specify one. Excludes that agent's own rule paths (since the agent reads those itself). Supported agents: cursor, opencode, copilot, claude, gemini, augment, windsurf, codex.
+    	Target agent to use (excludes rules from that agent's own paths). Supported agents: cursor, opencode, copilot, claude, gemini, augment, windsurf, codex.
   -w	Write rules to agent's config file and output only task to stdout. Requires agent (via task or -a flag).
 ```
 
@@ -179,14 +189,15 @@ Each of these would have a corresponding `.md` file (e.g., `triage-bug.md`, `fix
 
 The tool assembles the context in the following order:
 
-1.  **Rule Files**: It searches a list of predefined locations for rule files (`.md` or `.mdc`). These locations include the current directory, ancestor directories, user's home directory, and system-wide directories.
+1.  **Rule Files**: It searches for rule files (`.md` or `.mdc`) in directories specified via `-d` flags and automatically-added working directory and home directory.
 2.  **Rule Bootstrap Scripts**: For each rule file found (e.g., `my-rule.md`), it looks for an executable script named `my-rule-bootstrap`. If found, it runs the script before processing the rule file. These scripts are meant for bootstrapping the environment (e.g., installing tools) and their output is sent to `stderr`, not into the main context.
 3.  **Filtering**: If `-s` (include) flag is used, it parses the YAML frontmatter of each rule file to decide whether to include it. Note that selectors can only match top-level YAML fields (e.g., `language: go`), not nested fields.
 4.  **Task Prompt**: It searches for a task file matching the filename (without `.md` extension). Tasks are matched by filename, not by `task_name` in frontmatter. If selectors are provided with `-s`, they are used to filter between multiple task files with the same filename.
 5.  **Task Bootstrap Script**: For the task file found (e.g., `fix-bug.md`), it looks for an executable script named `fix-bug-bootstrap`. If found, it runs the script before processing the task file. This allows task-specific environment setup or data preparation.
-6.  **Parameter Expansion**: It substitutes variables in the task prompt using the `-p` flags.
-7.  **Output**: It prints the content of all included rule files, followed by the expanded task prompt, to standard output.
-8.  **Token Count**: A running total of estimated tokens is printed to standard error.
+6.  **User Prompt Appending**: If a user-prompt argument is provided, it is appended to the task content after a delimiter (`---`).
+7.  **Parameter Expansion**: It substitutes variables in the task prompt and user-prompt using the `-p` flags.
+8.  **Output**: It prints the content of all included rule files, followed by the expanded task prompt, to standard output.
+9.  **Token Count**: A running total of estimated tokens is printed to standard error.
 
 ### File Search Paths
 
@@ -194,7 +205,6 @@ The tool looks for task and rule files in the following locations, in order of p
 
 **Tasks:**
 - `./.agents/tasks/*.md` (task name matches filename without `.md` extension)
-- `~/.agents/tasks/*.md`
 
 **Commands** (reusable content blocks referenced via slash commands like `/command-name` inside task content):
 - `./.agents/commands/*.md`
@@ -204,9 +214,9 @@ The tool looks for task and rule files in the following locations, in order of p
 **Rules:**
 The tool searches for a variety of files and directories, including:
 - `CLAUDE.local.md`
-- `.agents/rules`, `.cursor/rules`, `.augment/rules`, `.windsurf/rules`, `.opencode/agent`, `.opencode/rules`
-- `.github/copilot-instructions.md`, `.github/agents`, `.gemini/styleguide.md`
-- `AGENTS.md`, `CLAUDE.md`, `GEMINI.md`, `.codex/AGENTS.md`
+- `.agents/rules`, `.cursor/rules`, `.augment/rules`, `.windsurf/rules`, `.opencode/agent`
+- `.github/copilot-instructions.md`, `.github/agents`, `.gemini/styleguide.md`, `.augment/guidelines.md`
+- `AGENTS.md`, `CLAUDE.md`, `GEMINI.md`, `.cursorrules`, `.windsurfrules`
 - User-specific rules in `~/.agents/rules`, `~/.claude/CLAUDE.md`, `~/.codex/AGENTS.md`, `~/.gemini/GEMINI.md`, `~/.opencode/rules`, etc.
 
 ### Remote File System Support
@@ -300,10 +310,10 @@ Deploy the application to production with all safety checks.
 You can then select the appropriate task using:
 ```bash
 # Deploy to staging
-coding-context -s environment=staging /deploy
+coding-context -s environment=staging deploy
 
 # Deploy to production
-coding-context -s environment=production /deploy
+coding-context -s environment=production deploy
 ```
 
 #### Task Frontmatter Selectors
@@ -328,11 +338,6 @@ When you run this task, it automatically applies the selectors:
 coding-context implement-feature
 ```
 
-This is equivalent to:
-```bash
-coding-context -s languages=go -s stage=implementation /implement-feature
-```
-
 **Selectors support OR logic for the same key using arrays:**
 ```markdown
 ---
@@ -394,13 +399,12 @@ This is particularly useful in agentic workflows where an AI agent has already b
 - Skipping all rules output
 
 **Example usage:**
-
 ```bash
 # Initial task invocation (includes all rules, uses task with resume: false)
-coding-context -s resume=false /fix-bug | ai-agent
+coding-context -s resume=false fix-bug | ai-agent
 
 # Resume the task (skips rules, uses task with resume: true)
-coding-context -r /fix-bug | ai-agent
+coding-context -r fix-bug | ai-agent
 ```
 
 **Example task files for resume mode:**
@@ -449,7 +453,7 @@ languages:
 To include this rule only when working on Go code, you would use `-s languages=go`:
 
 ```bash
-coding-context -s languages=go /fix-bug
+coding-context -s languages=go fix-bug
 ```
 
 This will include all rules with `languages: [ go ]` in their frontmatter, excluding rules for other languages.
@@ -468,10 +472,10 @@ Then select only the relevant rules:
 
 ```bash
 # Work on Python code with Python-specific rules
-coding-context -s languages=python /fix-bug
+coding-context -s languages=python fix-bug
 
 # Work on JavaScript code with JavaScript-specific rules
-coding-context -s languages=javascript /enhance-feature
+coding-context -s languages=javascript enhance-feature
 ```
 
 **Language Values**
@@ -503,42 +507,34 @@ If you need to filter on nested data, flatten your frontmatter structure to use
 
 ### Targeting a Specific Agent
 
-When working with a specific AI coding agent, the agent itself will read its own configuration files. The `-a` flag lets you specify which agent you're using, automatically excluding that agent's specific rule paths while including rules from other agents and generic rules.
+The `-a` flag specifies which AI coding agent you're using. This information is currently used for:
+
+1. **Write Rules Mode**: With the `-w` flag, determines where to write rules (e.g., `~/.github/agents/AGENTS.md` for `copilot`)
+
+> **Note:** Agent-based rule filtering is not currently implemented. All rules are included regardless of the `-a` value.
 
 **Supported agents:**
-- `cursor` - Excludes `.cursor/rules`, `.cursorrules`; includes other agents and generic rules
-- `opencode` - Excludes `.opencode/agent`, `.opencode/command`; includes other agents and generic rules
-- `copilot` - Excludes `.github/copilot-instructions.md`, `.github/agents`; includes other agents and generic rules
-- `claude` - Excludes `.claude/`, `CLAUDE.md`, `CLAUDE.local.md`; includes other agents and generic rules
-- `gemini` - Excludes `.gemini/`, `GEMINI.md`; includes other agents and generic rules
-- `augment` - Excludes `.augment/`; includes other agents and generic rules
-- `windsurf` - Excludes `.windsurf/`, `.windsurfrules`; includes other agents and generic rules
-- `codex` - Excludes `.codex/`, `AGENTS.md`; includes other agents and generic rules
+- `cursor` - Cursor IDE
+- `opencode` - OpenCode.ai  
+- `copilot` - GitHub Copilot
+- `claude` - Anthropic Claude
+- `gemini` - Google Gemini
+- `augment` - Augment
+- `windsurf` - Windsurf
+- `codex` - Codex
 
-**Example: Using Cursor:**
+**Example:**
 
 ```bash
-# When using Cursor, exclude .cursor/ and .cursorrules (Cursor reads those itself)
-# But include rules from other agents and generic rules
-coding-context -a cursor /fix-bug
+# Use with write rules mode
+coding-context -a copilot -w fix-bug
 ```
 
 **How it works:**
-- The `-a` flag sets the target agent
-- The target agent's own paths are excluded (e.g., `.cursor/` for cursor)
-- Rules from other agents are included (e.g., `.opencode/`, `.github/copilot-instructions.md`)
-- Generic rules (from `.agents/rules`) are always included
-- The agent name is automatically added as a selector, so generic rules can filter themselves with `agent: cursor` in frontmatter
-
-**Example generic rule with agent filtering:**
-
-```markdown
----
-agent: cursor
----
-# This rule only applies when using Cursor
-Use Cursor-specific features...
-```
+- The `-a` flag sets the target agent value
+- The agent value is stored in the context for use by the `-w` flag
+- With `-w`, the agent determines where to write rules to the user's home directory
+- All rules are currently included regardless of agent value
 
 **Agent field in task frontmatter:**
 
@@ -551,15 +547,7 @@ agent: cursor
 # This task automatically sets the agent to cursor
 ```
 
-This is useful for tasks designed for specific agents, ensuring the correct agent context is used regardless of command-line flags.
-
-**Use cases:**
-- **Avoid duplication**: The agent reads its own config, so exclude it from the context
-- **Cross-agent rules**: Include rules from other agents that might be relevant
-- **Generic rules**: Always include generic rules, with optional agent-specific filtering
-- **Task-specific agents**: Tasks can enforce a specific agent context
-
-The exclusion happens before rule processing, so excluded paths are never loaded or counted toward token estimates.
+This is useful for tasks designed for specific agents, ensuring the correct agent is set for determining the write path with `-w`.
 
 ### Bootstrap Scripts
 
diff --git a/docs/how-to/create-tasks.md b/docs/how-to/create-tasks.md
index de12dbd..b88e8d5 100644
--- a/docs/how-to/create-tasks.md
+++ b/docs/how-to/create-tasks.md
@@ -22,7 +22,7 @@ Please review the code changes with focus on:
 - Security implications
 ```
 
-Save as `.agents/tasks/code-review.md` (or `.agents/commands/code-review.md`).
+Save as `.agents/tasks/code-review.md`.
 
 Use with:
 ```bash
@@ -51,7 +51,7 @@ coding-context \
   -p feature_name="User Authentication" \
   -p requirements="OAuth2 support, secure password storage" \
   -p success_criteria="All tests pass, security audit clean" \
-  /implement-feature
+  implement-feature
 ```
 
 ## Multiple Tasks with Selectors
diff --git a/docs/how-to/github-actions.md b/docs/how-to/github-actions.md
index 901658a..0aa8fc8 100644
--- a/docs/how-to/github-actions.md
+++ b/docs/how-to/github-actions.md
@@ -209,7 +209,7 @@ Use the `-C` flag to run from a different directory:
 ```yaml
 - name: Assemble Context
   run: |
-    coding-context -C ./backend -s languages=go /fix-bug > context.txt
+    coding-context -C ./backend -s languages=go fix-bug > context.txt
 ```
 
 ## Best Practices
diff --git a/docs/index.md b/docs/index.md
index 2fcf929..f6bc387 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -19,7 +19,7 @@ sudo curl -fsL -o /usr/local/bin/coding-context \
 sudo chmod +x /usr/local/bin/coding-context
 
 # Use with an AI agent
-coding-context -p issue_key=BUG-123 -s languages=go /fix-bug | llm -m claude-3-5-sonnet-20241022
+coding-context -p issue_key=BUG-123 -s languages=go fix-bug | llm -m claude-3-5-sonnet-20241022
 ```
 
 ## Documentation Structure
diff --git a/docs/reference/cli.md b/docs/reference/cli.md
index 1622ccf..9cf9398 100644
--- a/docs/reference/cli.md
+++ b/docs/reference/cli.md
@@ -12,7 +12,7 @@ Complete reference for the `coding-context` command-line interface.
 ## Synopsis
 
 ```
-coding-context [options] <task-name>
+coding-context [options] <task-name> [user-prompt]
 ```
 
 ## Description
@@ -27,13 +27,27 @@ The Coding Context CLI assembles context from rule files and task prompts, perfo
 
 Task files can contain slash commands (e.g., `/command-name arg`) which reference command files for modular content reuse.
 
+### `[user-prompt]` (optional)
+
+**Optional.** Additional text to append to the task content. This text is appended after a delimiter (`---`) and can contain:
+- Slash commands (e.g., `/command-name arg`) which will be expanded
+- Parameter substitution placeholders (e.g., `${param}`)
+
+The user-prompt is processed the same way as task file content, allowing you to dynamically extend the task at runtime.
+
 **Examples:**
 ```bash
-# Task name (looks up fix-bug.md task file)
+# Task name only (looks up fix-bug.md task file)
 coding-context fix-bug
 
-# With parameters
-coding-context -p issue_key=BUG-123 fix-bug
+# Task with user-prompt
+coding-context fix-bug "Focus on the authentication module"
+
+# User-prompt with parameters
+coding-context -p issue_key=BUG-123 fix-bug "Check the error logs in /var/log"
+
+# User-prompt with slash commands
+coding-context fix-bug "/pre-checks and then analyze the code"
 
 # With selectors
 coding-context -s languages=go fix-bug
@@ -173,23 +187,27 @@ coding-context \
 **Type:** String  
 **Default:** (empty)
 
-Specify the target agent being used. When set, this excludes that agent's own rule paths (since the agent reads those itself) while including rules from other agents and generic rules.
+Specify the target agent being used. This is currently used for:
+1. **Write Rules Mode**: With `-w` flag, determines where to write rules (e.g., `~/.github/agents/AGENTS.md` for copilot)
+
+> **Note:** Agent-based rule filtering is not currently implemented. All rules are included regardless of the `-a` value.
 
 **Supported agents:** `cursor`, `opencode`, `copilot`, `claude`, `gemini`, `augment`, `windsurf`, `codex`
 
 **How it works:**
-- When `-a cursor` is specified, paths like `.cursor/rules` and `.cursorrules` are excluded
-- Rules from other agents (e.g., `.opencode/agent`, `.github/copilot-instructions.md`) are included
-- Generic rules from `.agents/rules` are always included
-- The agent name is automatically added as a selector for rule filtering
+- The agent value is stored in the context (can come from `-a` flag or task frontmatter)
+- With `-w` flag, the agent determines the user rules path for writing
+- All rules are currently included regardless of agent value
+
+**Agent Precedence:**
+- If a task specifies an `agent` field in its frontmatter, that takes precedence over the `-a` flag
+- The `-a` flag is used when the task doesn't specify an agent
+- Either the task's agent field or `-a` flag can be used to set the agent
 
 **Example:**
 ```bash
-# Using Cursor - excludes .cursor/ paths, includes others
-coding-context -a cursor fix-bug
-
-# Using GitHub Copilot - excludes .github/copilot-instructions.md, includes others
-coding-context -a copilot implement-feature
+# Use with write rules mode
+coding-context -a copilot -w implement-feature
 ```
 
 **Note:** Task files can override this with an `agent` field in their frontmatter.
@@ -240,40 +258,6 @@ coding-context -s languages=go -s priority=high fix-bug
 coding-context -s environment=production deploy
 ```
 
-**Note:** When filtering by language, use `-s languages=go` (plural). The selector key is `languages` (plural), matching the frontmatter field name.
-
-### `-a <agent>`
-
-**Type:** String (agent name)  
-**Default:** (empty)
-
-Specify the default agent to use. This acts as a fallback if the task doesn't specify an agent in its frontmatter.
-
-**Supported agents:**
-- `cursor` - [Cursor](https://cursor.sh/)
-- `opencode` - [OpenCode.ai](https://opencode.ai/)
-- `copilot` - [GitHub Copilot](https://github.com/features/copilot)
-- `claude` - [Anthropic Claude](https://claude.ai/)
-- `gemini` - [Google Gemini](https://gemini.google.com/)
-- `augment` - [Augment](https://augmentcode.com/)
-- `windsurf` - [Windsurf](https://codeium.com/windsurf)
-- `codex` - [Codex](https://codex.ai/)
-
-**Agent Precedence:**
-- If the task specifies an `agent` field in its frontmatter, that agent **overrides** the `-a` flag
-- The `-a` flag serves as a **default** agent when the task doesn't specify one
-- This allows tasks to specify their preferred agent while supporting a command-line default
-
-**Examples:**
-```bash
-# Use copilot as the default agent
-coding-context -a copilot fix-bug
-
-# Task with agent field will override -a flag
-# If fix-bug.md has "agent: claude", it will use claude instead of copilot
-coding-context -a copilot fix-bug
-```
-
 ### `-w`
 
 **Type:** Boolean flag  
@@ -372,15 +356,18 @@ coding-context fix-bug  # Bootstrap scripts can use these variables
 ### Basic Usage
 
 ```bash
-# Free-text prompt (used directly as task content)
-coding-context "Please help me review this code for security issues"
-
 # Task name lookup
 coding-context code-review
 
+# Task with user-prompt
+coding-context code-review "Focus on security vulnerabilities"
+
 # With parameters
 coding-context -p pr_number=123 code-review
 
+# User-prompt with parameters
+coding-context -p issue=BUG-456 fix-bug "Check the database connection logic"
+
 # With selectors
 coding-context -s languages=python fix-bug
 
@@ -390,6 +377,9 @@ coding-context \
   -s stage=implementation \
   -p feature_name="Authentication" \
   implement-feature
+
+# User-prompt with slash commands
+coding-context implement-feature "/pre-checks and validate the requirements"
 ```
 
 ### Working Directory
@@ -440,6 +430,9 @@ coding-context -r implement-feature | ai-agent
 # Claude
 coding-context fix-bug | claude
 
+# With user-prompt
+coding-context fix-bug "Focus on edge cases" | claude
+
 # LLM tool
 coding-context fix-bug | llm -m claude-3-5-sonnet-20241022
 
@@ -450,8 +443,8 @@ coding-context code-review | openai api completions.create -m gpt-4
 coding-context fix-bug > context.txt
 cat context.txt | your-ai-agent
 
-# Free-text prompt
-coding-context "Please help me debug the auth module" | claude
+# User-prompt with parameters
+coding-context -p issue=123 fix-bug "Check logs in /var/log" | claude
 ```
 
 ### Token Monitoring
diff --git a/docs/reference/file-formats.md b/docs/reference/file-formats.md
index 807240e..c1ad24b 100644
--- a/docs/reference/file-formats.md
+++ b/docs/reference/file-formats.md
@@ -442,13 +442,10 @@ This prevents command injection where expanded content could trigger further, un
 
 ### File Location
 
-Task files must be in one of these directories:
-- `./.agents/tasks/`
-- `./.cursor/commands/`
-- `./.opencode/command/`
-- `~/.agents/tasks/`
+Task files must be in these directories within any search path directory:
+- `.agents/tasks/`
 
-Tasks are matched by filename (without `.md` extension). The `task_name` field in frontmatter is optional and used only for metadata. For example, a file named `fix-bug.md` is matched by the command `/fix-bug`, regardless of whether it has `task_name` in its frontmatter.
+Tasks are matched by filename (without `.md` extension). The `task_name` field in frontmatter is optional and used only for metadata. For example, a file named `fix-bug.md` is matched by the task name `fix-bug`, regardless of whether it has `task_name` in its frontmatter.
 
 ## Command Files
 
@@ -523,11 +520,10 @@ Commands can also receive inline parameters:
 
 ### File Locations
 
-Command files must be in one of these directories:
-- `./.agents/commands/`
-- `./.cursor/commands/`
-- `./.opencode/command/`
-- `~/.agents/commands/`
+Command files must be in these directories within any search path directory:
+- `.agents/commands/`
+- `.cursor/commands/`
+- `.opencode/command/`
 
 Commands are matched by filename (without `.md` extension). For example, a file named `deploy.md` is matched by the slash command `/deploy`.
 
diff --git a/docs/reference/search-paths.md b/docs/reference/search-paths.md
index b354675..0ffd7c6 100644
--- a/docs/reference/search-paths.md
+++ b/docs/reference/search-paths.md
@@ -81,10 +81,7 @@ Rule files are discovered from directories specified via the `-d` flag (plus aut
 .augment/rules/
 .windsurf/rules/
 .opencode/agent/
-.opencode/command/
-.opencode/rules/
 .github/agents/
-.codex/
 ```
 
 **Specific files:**
@@ -92,6 +89,7 @@ Rule files are discovered from directories specified via the `-d` flag (plus aut
 CLAUDE.local.md
 .github/copilot-instructions.md
 .gemini/styleguide.md
+.augment/guidelines.md
 ```
 
 **Standard files:**
diff --git a/main.go b/main.go
index 9004881..58e4279 100644
--- a/main.go
+++ b/main.go
@@ -34,8 +34,8 @@ func main() {
 
 	flag.StringVar(&workDir, "C", ".", "Change to directory before doing anything.")
 	flag.BoolVar(&resume, "r", false, "Resume mode: skip outputting rules and select task with 'resume: true' in frontmatter.")
-	flag.BoolVar(&writeRules, "w", false, "Write rules to the agent's user rules path and only print the prompt to stdout. Requires -a flag.")
-	flag.Var(&agent, "a", "Target agent to use (excludes rules from other agents). Supported agents: cursor, opencode, copilot, claude, gemini, augment, windsurf, codex.")
+	flag.BoolVar(&writeRules, "w", false, "Write rules to the agent's user rules path and only print the prompt to stdout. Requires agent (via task 'agent' field or -a flag).")
+	flag.Var(&agent, "a", "Target agent to use. Required when using -w to write rules to the agent's user rules path. Supported agents: cursor, opencode, copilot, claude, gemini, augment, windsurf, codex.")
 	flag.Var(&params, "p", "Parameter to substitute in the prompt. Can be specified multiple times as key=value.")
 	flag.Var(&includes, "s", "Include rules with matching frontmatter. Can be specified multiple times as key=value.")
 	flag.Func("d", "Directory containing rules and tasks. Can be specified multiple times. Supports various protocols via go-getter (http://, https://, git::, s3::, file:// etc.).", func(s string) error {