@nano-step/skill-manager 4.0.0 → 5.0.1

package/{templates/skill → skills/skill-management}/references/error-handling.md

@@ -12,7 +12,7 @@
 
 ## Automatic Retry Mechanism
 
-The agent-skill-manager automatically retries failed tool executions before returning errors.
+The skill-manager automatically retries failed tool executions before returning errors.
 
 ### Retry Configuration
 - Maximum attempts: 3
@@ -67,7 +67,7 @@ Template:
 ```
 Error: Tool not found: <tool-id>
 Suggestions: <tool-a>, <tool-b>, <tool-c>
-Next: confirm the intended tool or run /agent-skill-refresh.
+Next: confirm the intended tool or run /skill-refresh.
 ```
 
 ## Execution Failure Handling
@@ -113,18 +113,18 @@ Next: confirm retry with new parameters
 
 ## Cache Missing Recovery
 Symptoms:
-- `.opencode/agent-skill-tools.json` missing.
+- `.opencode/skill-tools.json` missing.
 - Cache file invalid or unreadable.
 
 Recovery steps:
 1. Inform that cache is missing or invalid.
-2. Recommend /agent-skill-refresh to regenerate.
+2. Recommend /skill-refresh to regenerate.
 3. Proceed with dynamic tool discovery if possible.
 
 Template:
 ```
-Cache missing or invalid at .opencode/agent-skill-tools.json
-Suggestion: run /agent-skill-refresh
+Cache missing or invalid at .opencode/skill-tools.json
+Suggestion: run /skill-refresh
 Fallback: dynamic discovery enabled
 ```
 
@@ -247,7 +247,7 @@ Decision tree (text form):
    - Yes -> request corrected params.
    - No -> continue.
 5. Is cache missing?
-   - Yes -> recommend /agent-skill-refresh and use dynamic discovery.
+   - Yes -> recommend /skill-refresh and use dynamic discovery.
    - No -> continue.
 6. Unknown failure
    - Provide generic error template and ask for clarification.

package/{templates/skill → skills/skill-management}/references/tool-categories.md

@@ -6,7 +6,7 @@
 
 This skill uses **semantic categorization**:
 
-1. **Tool discovery** - `/agent-skill-refresh` enumerates all tools available to the agent
+1. **Tool discovery** - `/skill-refresh` enumerates all tools available to the agent
 2. **Semantic analysis** - the AI reads each tool's **name + description**
 3. **Dynamic grouping** - tools are grouped into categories based on what they do
 
@@ -31,12 +31,12 @@ categories are **labels**, not fixed contracts, and can evolve with the tool set
 
 - Categories are **not hard-coded**
 - Category names are **derived from the tools** at refresh time
-- Custom MCP servers are handled automatically
-- The system works with any MCP configuration (MetaMCP, standalone, or custom)
+- Custom tool servers are handled automatically
+- The system works with any tool configuration (MetaMCP, standalone, or custom)
 
 ## Cache File (v2.0.0) - How to Read It
 
-The `/agent-skill-refresh` command writes a cache file using the v2.0.0 schema. This
+The `/skill-refresh` command writes a cache file using the v2.0.0 schema. This
 file is the **source of truth** for the current categorization state.
 
 ### Schema Overview

package/{templates/skill → skills/skill-management}/references/tool-execution.md

@@ -3,11 +3,11 @@
 # Do not include category catalogs, result summarization, or error templates here.
 
 ## Cache Access and Reading
-Cache location: .opencode/agent-skill-tools.json
-Purpose: fast tool routing without reloading full MCP schemas.
+Cache location: .opencode/skill-tools.json
+Purpose: fast tool routing without reloading full tool schemas.
 
 How to read the cache:
-1. Check if file exists at project root `.opencode/agent-skill-tools.json`.
+1. Check if file exists at project root `.opencode/skill-tools.json`.
 2. Parse JSON into a structured object.
 3. Validate minimal fields: version, refreshed_at, tool_count, categories.
 4. If missing or malformed, fall back to dynamic discovery.
@@ -18,10 +18,10 @@ Cache staleness heuristic:
 
 ## Cache Schema (TypeScript Interface)
 ```ts
-interface AgentSkillToolCache {
+interface SkillToolCache {
   version: string;
   refreshed_at: string; // ISO string
-  mcp_servers: string[];
+  tool_sources: string[];
   tool_count: number;
   categories: {
     [name: string]: {
@@ -156,7 +156,7 @@ params: { fullPage: true }
 ## When to Avoid Cache
 - During tool upgrades or when discrepancies are reported.
 - When a tool id from cache fails execution.
-- When the user references a new MCP server prefix.
+- When the user references a new tool server prefix.
 
 ## Execution Prerequisites
 - Ensure selected tool exists in cache or registry.
@@ -234,7 +234,7 @@ Tool: sequentialthinking { thought: "...", thoughtNumber: 1, totalThoughts: 4, n
 ## Cache Refresh Recommendation Text
 Recommended message:
 ```
-Cache appears stale or missing. Run /agent-skill-refresh to rebuild tool metadata.
+Cache appears stale or missing. Run /skill-refresh to rebuild tool metadata.
 ```
 
 ## Execution Output Metadata (Internal)

package/{templates/skill → skills/skill-management}/references/workflows.md

@@ -70,7 +70,7 @@ Each prerequisite step has:
 
 ## Session State
 
-Completed prerequisites are tracked in `.opencode/.agent-skill-session.json`:
+Completed prerequisites are tracked in `.opencode/.skill-session.json`:
 
 ```json
 {
@@ -202,26 +202,26 @@ Also matched: 'custom-db-workflow'
 
 ## Managing Workflows
 
-Use `/agent-skill-workflow` command:
+Use `/skill-workflow` command:
 
 ```bash
 # List all workflows
-/agent-skill-workflow list
+/skill-workflow list
 
 # Add from template
-/agent-skill-workflow add --template database
+/skill-workflow add --template database
 
 # Add custom workflow
-/agent-skill-workflow add my-workflow
+/skill-workflow add my-workflow
 
 # Edit existing
-/agent-skill-workflow edit my-workflow
+/skill-workflow edit my-workflow
 
 # Disable temporarily
-/agent-skill-workflow disable my-workflow
+/skill-workflow disable my-workflow
 
 # Remove
-/agent-skill-workflow remove my-workflow
+/skill-workflow remove my-workflow
 ```
 
 ## Best Practices
@@ -230,4 +230,4 @@ Use `/agent-skill-workflow` command:
 2. **Use enforce sparingly** - Only for critical safety workflows
 3. **Keep prerequisites minimal** - Each step adds latency
 4. **Mark optional steps** - Use `required: false` for nice-to-have steps
-5. **Review session state** - Check `.agent-skill-session.json` if prerequisites seem stuck
+5. **Review session state** - Check `.skill-session.json` if prerequisites seem stuck

package/{templates/command-refresh.md → skills/skill-management/skill-refresh.md}

@@ -1,10 +1,10 @@
-# /agent-skill-refresh
+# /skill-refresh
 
-Re-index and categorize all available MCP tools for intelligent routing.
+Re-index and categorize all available tools for intelligent routing.
 
 ## Purpose
 
-Create a semantic tool index in `.opencode/agent-skill-tools.json` by analyzing **all available tools** in the current agent context. Categorization is AI-only and based on tool names + descriptions (no prefixes).
+Create a semantic tool index in `.opencode/skill-tools.json` by analyzing **all available tools** in the current agent context. Categorization is AI-only and based on tool names + descriptions (no prefixes).
 
 ## When to Run
 
@@ -15,10 +15,10 @@ Create a semantic tool index in `.opencode/agent-skill-tools.json` by analyzing
 
 ## Auto-Refresh Triggers
 
-The agent-skill-manager will suggest running `/agent-skill-refresh` automatically when:
+The skill-manager will suggest running `/skill-refresh` automatically when:
 
 - **Tool not found**: A requested tool doesn't exist in the cache
-- **Cache missing**: The `.opencode/agent-skill-tools.json` file doesn't exist
+- **Cache missing**: The `.opencode/skill-tools.json` file doesn't exist
 - **Cache stale**: The cache is older than 24 hours
 - **Tool count mismatch**: Available tools differ from cached count
 
@@ -59,7 +59,7 @@ Output as JSON.
 
 ### Step 3: Write Cache File (v2.0.0)
 
-Write `.opencode/agent-skill-tools.json` with this schema:
+Write `.opencode/skill-tools.json` with this schema:
 
 ```json
 {
@@ -86,7 +86,7 @@ Write `.opencode/agent-skill-tools.json` with this schema:
 ### Step 4: Report Summary
 
 ```
-Agent Skill Tools Index Refreshed:
+Skill Tools Index Refreshed:
 
 Categories:
   - browser-automation: 25 tools
@@ -96,5 +96,5 @@ Categories:
 Summary:
   - Total tools: 75
   - Categories: 3
-  - Cache: .opencode/agent-skill-tools.json
+  - Cache: .opencode/skill-tools.json
 ```

package/skills/skill-management/skill.json

@@ -0,0 +1,18 @@
+{
+  "name": "skill-management",
+  "version": "1.0.0",
+  "description": "AI skill routing and execution — isolates tool definitions in subagent context to save 80-95% tokens",
+  "compatibility": "OpenCode with any MCP server",
+  "agent": {
+    "id": "skill-manager",
+    "config": {
+      "mode": "subagent",
+      "model": "gitlab/duo-chat-haiku-4-5",
+      "temperature": 0.1,
+      "description": "Skill router - isolates tool definitions in separate context window to save tokens",
+      "prompt_append": "You are skill-manager, a specialized agent for routing and executing tools. Your purpose is to ISOLATE tool definitions from the main agent's context window, reducing token usage by 80-95%.\n\nWORKFLOW:\n1. Read .opencode/skill-tools.json cache for tool routing (if exists)\n2. Parse the task to identify which skill category is needed (browser, github, graphql, docs, reasoning)\n3. Select the appropriate tool(s) from that category\n4. Execute the tool with correct parameters\n5. Filter/summarize results before returning to main agent\n\nTOOL CATEGORIES:\n- browser: chrome-devtools__* (screenshots, clicks, navigation, DOM)\n- github: github-*__* (PRs, issues, repos, commits, branches)\n- graphql: graphql-tools__* (schema, queries, mutations)\n- docs: context7__* (documentation lookup, library references)\n- reasoning: Sequential-Thinking__* (complex multi-step analysis)\n\nRESULT HANDLING:\n- For large results (>1000 chars): Summarize key information\n- For file operations: Return path/status, not full content\n- For data queries: Return relevant subset, not full dataset\n- Always include: Success/failure status, key identifiers, actionable next steps\n\nERROR HANDLING:\n- Tool not found: Suggest similar tools or ask for clarification\n- Tool execution failed: Return error details and potential fixes\n- Cache missing: Prompt user to run /skill-refresh command\n- Never silently fail\n\nYou are FAST (Haiku model) and DETERMINISTIC (temp 0.1). Focus on accurate tool selection and execution."
+    }
+  },
+  "commands": ["skill-refresh.md"],
+  "tags": ["skill", "routing", "token-saving"]
+}

package/dist/install.d.ts

@@ -1 +0,0 @@
-export declare function install(): Promise<void>;

package/dist/install.js

@@ -1,35 +0,0 @@
-"use strict";
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.install = install;
-const path_1 = __importDefault(require("path"));
-const chalk_1 = __importDefault(require("chalk"));
-const fs_extra_1 = __importDefault(require("fs-extra"));
-const utils_1 = require("./utils");
-async function install() {
-    const paths = await (0, utils_1.detectOpenCodePaths)();
-    await (0, utils_1.ensureDirExists)(paths.commandDir);
-    await (0, utils_1.ensureDirExists)(paths.skillsDir);
-    const skillTargetDir = path_1.default.join(paths.skillsDir, utils_1.SKILL_DIR_NAME);
-    const commandTargetPath = path_1.default.join(paths.commandDir, "agent-skill-refresh.md");
-    await fs_extra_1.default.copy(paths.templateSkillDir, skillTargetDir, { overwrite: true });
-    const commandTemplate = await (0, utils_1.readText)(paths.templateCommandPath);
-    await (0, utils_1.writeText)(commandTargetPath, commandTemplate);
-    const agentTemplate = await (0, utils_1.readJsonFile)(paths.templateAgentPath, {});
-    const agentConfig = await (0, utils_1.readJsonFile)(paths.agentConfigPath, {});
-    const agents = (agentConfig.agents || {});
-    const templateAgents = (agentTemplate.agents || agentTemplate);
-    agentConfig.agents = { ...agents, ...templateAgents };
-    await (0, utils_1.writeJsonFile)(paths.agentConfigPath, agentConfig);
-    await (0, utils_1.writeJsonFile)(paths.versionFilePath, { version: utils_1.PACKAGE_VERSION, installedAt: new Date().toISOString() });
-    const state = await (0, utils_1.getInstallationState)(paths);
-    if (!state.skillInstalled || !state.commandInstalled || !state.agentInstalled) {
-        throw new Error("Installation verification failed. Some files were not installed.");
-    }
-    if (Object.keys(agentConfig).length > 0 && Object.prototype.hasOwnProperty.call(agentConfig, utils_1.AGENT_ID)) {
-        console.log(chalk_1.default.yellow("agent-skill-manager agent already existed; configuration was updated."));
-    }
-    console.log(chalk_1.default.green("Agent skill manager installed successfully."));
-}

package/dist/remove.d.ts

@@ -1 +0,0 @@
-export declare function remove(): Promise<void>;

package/dist/remove.js

@@ -1,32 +0,0 @@
-"use strict";
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.remove = remove;
-const path_1 = __importDefault(require("path"));
-const chalk_1 = __importDefault(require("chalk"));
-const fs_extra_1 = __importDefault(require("fs-extra"));
-const utils_1 = require("./utils");
-async function remove() {
-    const paths = await (0, utils_1.detectOpenCodePaths)();
-    const skillTargetDir = path_1.default.join(paths.skillsDir, utils_1.SKILL_DIR_NAME);
-    const commandTargetPath = path_1.default.join(paths.commandDir, "agent-skill-refresh.md");
-    if (await fs_extra_1.default.pathExists(skillTargetDir)) {
-        await fs_extra_1.default.remove(skillTargetDir);
-    }
-    if (await fs_extra_1.default.pathExists(commandTargetPath)) {
-        await fs_extra_1.default.remove(commandTargetPath);
-    }
-    const agentConfig = await (0, utils_1.readJsonFile)(paths.agentConfigPath, {});
-    const agents = (agentConfig.agents || {});
-    if (Object.prototype.hasOwnProperty.call(agents, utils_1.AGENT_ID)) {
-        delete agents[utils_1.AGENT_ID];
-        agentConfig.agents = agents;
-        await (0, utils_1.writeJsonFile)(paths.agentConfigPath, agentConfig);
-    }
-    if (await fs_extra_1.default.pathExists(paths.versionFilePath)) {
-        await fs_extra_1.default.remove(paths.versionFilePath);
-    }
-    console.log(chalk_1.default.green("Agent skill manager removed successfully."));
-}

package/dist/update.d.ts

@@ -1 +0,0 @@
-export declare function update(): Promise<void>;

package/dist/update.js

@@ -1,53 +0,0 @@
-"use strict";
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-exports.update = update;
-const path_1 = __importDefault(require("path"));
-const chalk_1 = __importDefault(require("chalk"));
-const fs_extra_1 = __importDefault(require("fs-extra"));
-const utils_1 = require("./utils");
-async function update() {
-    const paths = await (0, utils_1.detectOpenCodePaths)();
-    await (0, utils_1.ensureDirExists)(paths.commandDir);
-    await (0, utils_1.ensureDirExists)(paths.skillsDir);
-    const skillTargetDir = path_1.default.join(paths.skillsDir, utils_1.SKILL_DIR_NAME);
-    const commandTargetPath = path_1.default.join(paths.commandDir, "agent-skill-refresh.md");
-    await (0, utils_1.ensureDirExists)(skillTargetDir);
-    const state = await (0, utils_1.getInstallationState)(paths);
-    if (state.installedVersion === utils_1.PACKAGE_VERSION) {
-        console.log(chalk_1.default.green("Agent skill manager is already up to date."));
-        return;
-    }
-    const [skillCustomized, commandCustomized] = await Promise.all([
-        (0, utils_1.directoryDiffersFromTemplate)(skillTargetDir, paths.templateSkillDir),
-        (0, utils_1.fileDiffersFromTemplate)(commandTargetPath, paths.templateCommandPath),
-    ]);
-    if (skillCustomized || commandCustomized) {
-        console.log(chalk_1.default.yellow("Detected customized files. Backups will be created before update."));
-    }
-    if (await fs_extra_1.default.pathExists(skillTargetDir)) {
-        console.log(chalk_1.default.yellow("Backing up existing skill directory before update."));
-        const timestamp = new Date().toISOString().replace(/[:.]/g, "-");
-        const backupDir = `${skillTargetDir}.backup-${timestamp}.bak`;
-        await fs_extra_1.default.copy(skillTargetDir, backupDir, { overwrite: true });
-    }
-    if (await fs_extra_1.default.pathExists(commandTargetPath)) {
-        await (0, utils_1.backupFile)(commandTargetPath, "backup");
-    }
-    if (await fs_extra_1.default.pathExists(paths.agentConfigPath)) {
-        await (0, utils_1.backupFile)(paths.agentConfigPath, "backup");
-    }
-    await fs_extra_1.default.copy(paths.templateSkillDir, skillTargetDir, { overwrite: true });
-    const commandTemplate = await fs_extra_1.default.readFile(paths.templateCommandPath, "utf8");
-    await fs_extra_1.default.writeFile(commandTargetPath, commandTemplate, "utf8");
-    const agentTemplate = await (0, utils_1.readJsonFile)(paths.templateAgentPath, {});
-    const agentConfig = await (0, utils_1.readJsonFile)(paths.agentConfigPath, {});
-    const agents = (agentConfig.agents || {});
-    const templateAgents = (agentTemplate.agents || agentTemplate);
-    agentConfig.agents = { ...agents, ...templateAgents };
-    await (0, utils_1.writeJsonFile)(paths.agentConfigPath, agentConfig);
-    await (0, utils_1.writeJsonFile)(paths.versionFilePath, { version: utils_1.PACKAGE_VERSION, updatedAt: new Date().toISOString() });
-    console.log(chalk_1.default.green("Agent skill manager updated successfully."));
-}

package/templates/agent.json

@@ -1,11 +0,0 @@
-{
-  "agents": {
-    "mcp-manager": {
-      "mode": "subagent",
-      "model": "gitlab/duo-chat-haiku-4-5",
-      "temperature": 0.1,
-      "description": "MCP tool router - isolates MCP tools in separate context window to save tokens",
-      "prompt_append": "You are mcp-manager, a specialized agent for executing MCP (Model Context Protocol) tools. Your purpose is to ISOLATE MCP tool definitions from the main agent's context window, reducing token usage by 80-95%.\n\nWORKFLOW:\n1. Read .opencode/mcp-tools.json cache for tool routing (if exists)\n2. Parse the task to identify which MCP category is needed (browser, github, graphql, docs, reasoning)\n3. Select the appropriate tool(s) from that category\n4. Execute the tool with correct parameters\n5. Filter/summarize results before returning to main agent\n\nTOOL CATEGORIES:\n- browser: chrome-devtools__* (screenshots, clicks, navigation, DOM)\n- github: github-*__* (PRs, issues, repos, commits, branches)\n- graphql: graphql-tools__* (schema, queries, mutations)\n- docs: context7__* (documentation lookup, library references)\n- reasoning: Sequential-Thinking__* (complex multi-step analysis)\n\nRESULT HANDLING:\n- For large results (>1000 chars): Summarize key information\n- For file operations: Return path/status, not full content\n- For data queries: Return relevant subset, not full dataset\n- Always include: Success/failure status, key identifiers, actionable next steps\n\nERROR HANDLING:\n- Tool not found: Suggest similar tools or ask for clarification\n- Tool execution failed: Return error details and potential fixes\n- Cache missing: Prompt user to run /mcp-refresh command\n- Never silently fail\n\nYou are FAST (Haiku model) and DETERMINISTIC (temp 0.1). Focus on accurate tool selection and execution."
-    }
-  }
-}

package/templates/command-workflow.md

@@ -1,188 +0,0 @@
-# /agent-skill-workflow
-
-Manage workflow rules that define prerequisite steps before tool execution.
-
-## Purpose
-
-Create, list, edit, and remove workflow definitions that automatically execute prerequisite steps when certain tools are triggered. For example, always inspect database structure before running queries.
-
-## Subcommands
-
-### List Workflows
-
-```
-/agent-skill-workflow list
-```
-
-Display all defined workflows with their status, triggers, and prerequisite count.
-
-**Output format:**
-```
-Workflows (3 defined):
-
-✓ database-safe-query [enforce]
-  Triggers: category=database, keywords=[query, select]
-  Prerequisites: 4 steps
-  
-✓ browser-safe-interaction [enforce]
-  Triggers: category=browser
-  Prerequisites: 1 step
-
-○ github-pr-review [disabled]
-  Triggers: tools=[merge_pull_request]
-  Prerequisites: 3 steps
-```
-
-### Add Workflow (Interactive)
-
-```
-/agent-skill-workflow add <name>
-```
-
-Create a new workflow interactively. The agent will prompt for:
-
-1. **Description**: What does this workflow do?
-2. **Triggers**: 
-   - Category (optional): Which tool category triggers this?
-   - Tools (optional): Which specific tools trigger this?
-   - Keywords (optional): Which words in the task trigger this?
-3. **Prerequisites**: For each step:
-   - Tool name
-   - Description (why this step is needed)
-   - Required? (yes/no)
-4. **Mode**: enforce, warn, or suggest
-
-**Example session:**
-```
-> /agent-skill-workflow add api-safety
-
-Description: Check API schema before making requests
-Trigger category (or skip): api
-Trigger tools (comma-separated, or skip): execute_request, post_request
-Trigger keywords (comma-separated, or skip): api call, request
-
-Prerequisite 1:
-  Tool: get_schema
-  Description: Fetch API schema
-  Required? yes
-  
-Add another prerequisite? yes
-
-Prerequisite 2:
-  Tool: list_endpoints
-  Description: List available endpoints
-  Required? no
-
-Add another prerequisite? no
-
-Mode (enforce/warn/suggest): warn
-
-✓ Workflow 'api-safety' created with 2 prerequisites
-```
-
-### Add from Template
-
-```
-/agent-skill-workflow add --template <template-name>
-```
-
-Create a workflow from a built-in template. Available templates:
-
-| Template | Description | Prerequisites |
-|----------|-------------|---------------|
-| `database` | Safe database queries | list_databases → list_tables → inspect_table → get_indexes |
-| `browser` | Safe browser interaction | take_snapshot |
-| `github-pr` | PR review workflow | get_pull_request → get_files → get_status |
-| `graphql` | Schema-aware GraphQL | get_schema → filter_types |
-
-**Example:**
-```
-> /agent-skill-workflow add --template database
-
-✓ Created workflow 'database-safe-query' from template
-  Mode: enforce
-  Prerequisites: 4 steps
-```
-
-To list available templates:
-```
-/agent-skill-workflow add --template
-```
-
-### Edit Workflow
-
-```
-/agent-skill-workflow edit <name>
-```
-
-Modify an existing workflow. Shows current values and allows changing any field.
-
-### Remove Workflow
-
-```
-/agent-skill-workflow remove <name>
-```
-
-Delete a workflow after confirmation.
-
-```
-> /agent-skill-workflow remove api-safety
-
-Remove workflow 'api-safety'? This cannot be undone. (yes/no): yes
-
-✓ Workflow 'api-safety' removed
-```
-
-### Enable/Disable Workflow
-
-```
-/agent-skill-workflow enable <name>
-/agent-skill-workflow disable <name>
-```
-
-Toggle a workflow's enabled status without removing it.
-
-```
-> /agent-skill-workflow disable database-safe-query
-
-✓ Workflow 'database-safe-query' disabled
-  (Prerequisites will not run until re-enabled)
-```
-
-## Workflow Modes
-
-| Mode | Behavior |
-|------|----------|
-| `enforce` | Prerequisites run automatically before main action |
-| `warn` | Warning shown, user can skip prerequisites |
-| `suggest` | Prerequisites mentioned but don't block execution |
-
-## Storage
-
-Workflows are stored in `.opencode/agent-skill-tools.json` under the `workflows` key. They persist across sessions and are preserved when running `/agent-skill-refresh`.
-
-## Session State
-
-Completed prerequisites are tracked in `.opencode/.agent-skill-session.json`. Within a session, prerequisites only run once per workflow (unless the session expires after 24 hours).
-
-## Examples
-
-### Create a custom workflow for file operations
-```
-/agent-skill-workflow add file-safety
-```
-Then define triggers for file tools and prerequisites like "list directory" before "delete file".
-
-### Use database template and customize
-```
-/agent-skill-workflow add --template database
-/agent-skill-workflow edit database-safe-query
-```
-Modify the default database workflow to fit your needs.
-
-### Temporarily disable a workflow
-```
-/agent-skill-workflow disable browser-safe-interaction
-# ... do some quick browser tasks without snapshots ...
-/agent-skill-workflow enable browser-safe-interaction
-```