claw-doing 0.1.0

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 seekcontext
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -0,0 +1,270 @@
+# claw-doing
+
+> See what your agent is doing — right now.
+
+**claw-doing** is an [OpenClaw](https://openclaws.io) plugin that gives long-running agent tasks real-time progress visibility. Instead of waiting silently for minutes while your agent works, you get a live step-by-step progress feed — right in your messaging channel.
+
+---
+
+## The problem
+
+When you ask OpenClaw to do something complex — research a topic, process files, build a multi-step pipeline — the interaction goes silent:
+
+```
+You:   "Research AI agent frameworks and write a report"
+Agent: [silent for 8 minutes]
+You:   "...is it still running?"
+Agent: [finally replies, or times out]
+```
+
+You have no idea what's happening, whether it's stuck, or when it will finish.
+
+## What claw-doing does
+
+With claw-doing installed, the same interaction looks like this:
+
+```
+You:   "Research AI agent frameworks and write a report"
+
+Agent: Research AI agent frameworks
+       Task ID: `a3f8c2d1-...`
+
+       Plan — 4 steps:
+         1. Search and collect sources (~3 min)
+         2. Extract key findings (~2 min)
+         3. Cross-reference and validate (~2 min)
+         4. Write and deliver report (~3 min)
+
+       [░░░░░░░░░░] 0/4
+       Starting now...
+
+       ──── (3 minutes later) ────
+
+       [███░░░░░░░] 1/4 — elapsed: 3m 12s
+
+       ✓ Step 1: Found 47 sources, filtered to 14 high-quality references
+       ⚠ Note: Two sources were paywalled — used abstracts only
+
+       Next: Step 2 — Extract key findings
+
+       ──── (2 minutes later) ────
+
+       [██████░░░░] 2/4 — elapsed: 5m 8s
+
+       ✓ Step 2: Extracted 31 key data points across 8 frameworks
+
+       ... and so on ...
+
+       [██████████] 4/4 — 11m 42s total
+
+       ✅ Completed successfully
+
+       Compiled a 2,800-word report covering 8 frameworks with 12 citations.
+       Key finding: LangGraph leads on multi-agent orchestration; OpenClaw leads on channel integrations.
+
+       Deliverables:
+         • memory/2026-03-29-ai-agent-frameworks.md
+```
+
+---
+
+## Installation
+
+```bash
+openclaw plugins install claw-doing
+```
+
+OpenClaw checks ClawHub first, then falls back to npm automatically.
+
+After installation, restart your OpenClaw gateway to load the plugin.
+
+### Enable the bootstrap instructions (recommended)
+
+The plugin ships with `bootstrap/TASK_TRACKER.md` — a set of instructions that teaches your agent when and how to use the tracking tools. Add it to your OpenClaw workspace bootstrap:
+
+```json
+{
+  "hooks": {
+    "internal": {
+      "enabled": true,
+      "entries": {
+        "bootstrap-extra-files": {
+          "enabled": true,
+          "paths": ["node_modules/claw-doing/bootstrap/TASK_TRACKER.md"]
+        }
+      }
+    }
+  }
+}
+```
+
+Or copy it manually to your workspace:
+
+```bash
+cp node_modules/claw-doing/bootstrap/TASK_TRACKER.md ~/.openclaw/workspace/TASK_TRACKER.md
+```
+
+Then add `TASK_TRACKER.md` to your bootstrap file list so the agent loads it on start.
+
+---
+
+## How it works
+
+claw-doing registers three agent tools. The agent calls them as it works — no separate dashboard or server required.
+
+### Tools
+
+#### `task_plan`
+
+Called at the **start** of a multi-step task. Registers the plan and returns a task ID.
+
+| Parameter | Type | Description |
+|-----------|------|-------------|
+| `title` | `string` | Short title for the task |
+| `steps` | `Step[]` | Ordered list of steps |
+| `steps[].index` | `number` | Step number (1-based) |
+| `steps[].description` | `string` | What this step does |
+| `steps[].estimatedMinutes` | `number?` | Optional time estimate |
+
+Returns: task ID + formatted plan for the user.
+
+#### `task_step_done`
+
+Called after **each step** completes. Updates progress and notifies the user.
+
+| Parameter | Type | Description |
+|-----------|------|-------------|
+| `taskId` | `string` | Task ID from `task_plan` |
+| `stepIndex` | `number` | Which step was just completed |
+| `summary` | `string` | What was accomplished |
+| `deviation` | `string?` | Unexpected findings (optional) |
+
+Returns: progress bar + step summary.
+
+#### `task_complete`
+
+Called when the **entire task** finishes (or fails). Delivers the final report.
+
+| Parameter | Type | Description |
+|-----------|------|-------------|
+| `taskId` | `string` | Task ID from `task_plan` |
+| `outcome` | `"success" \| "partial" \| "failed"` | Task outcome |
+| `summary` | `string` | User-facing completion summary |
+| `artifacts` | `string[]?` | Output files or results produced |
+
+Returns: final progress bar + summary + deliverables list.
+
+### Task state
+
+Each task is persisted as a JSON file at:
+
+```
+~/.openclaw/workspace/claw-doing/tasks/<taskId>.json
+```
+
+This means task state survives agent restarts and can be inspected or archived.
+
+---
+
+## Configuration
+
+Optional config in your `openclaw.config.json`:
+
+```json
+{
+  "plugins": {
+    "claw-doing": {
+      "storageDir": "/custom/path/to/tasks",
+      "progressBarWidth": 12
+    }
+  }
+}
+```
+
+| Option | Default | Description |
+|--------|---------|-------------|
+| `storageDir` | `~/.openclaw/workspace/claw-doing/tasks/` | Where task JSON files are stored |
+| `progressBarWidth` | `10` | Width of the `█░` progress bar |
+
+You can also set the storage directory via environment variable:
+
+```bash
+CLAW_DOING_STORAGE_DIR=/my/custom/path openclaw start
+```
+
+---
+
+## Design principles
+
+**Zero deployment.** Unlike dashboard-based solutions, claw-doing requires no external server, database, or Docker container. Install the plugin and it works.
+
+**In-channel delivery.** Progress updates arrive in the same messaging channel you're already using — Discord, Telegram, WhatsApp, web chat. You don't need to open a separate UI.
+
+**Agent-driven, not hook-driven.** The agent explicitly calls `task_step_done` after each step. This makes progress reporting intentional and semantically rich (the agent decides what counts as "done" and what to say about it), rather than relying on low-level hook instrumentation that may fire at the wrong granularity.
+
+**Persistent state.** Task records are written to disk as JSON. They can be read, archived, or used by other tooling (dashboards, scripts) independently.
+
+---
+
+## Comparison with similar projects
+
+| | claw-doing | ClawBeacon |
+|---|---|---|
+| **Form factor** | OpenClaw plugin (zero deployment) | Separate web app (Docker/Railway) |
+| **Where you see progress** | Your chat channel | A Kanban dashboard in a browser |
+| **Setup** | `openclaw plugins install claw-doing` | Clone, deploy backend + frontend |
+| **Best for** | Individual users, any task | Teams managing multiple agents |
+
+---
+
+## Development
+
+```bash
+git clone https://github.com/seekcontext/claw-doing
+cd claw-doing
+npm install
+npm run dev        # watch mode
+npm run typecheck  # type check without emitting
+npm run build      # compile to dist/
+```
+
+### Project structure
+
+```
+claw-doing/
+├── index.ts                      # Plugin entry — registers all tools
+├── src/
+│   ├── types.ts                  # Shared Task and TaskStep interfaces
+│   ├── store/
+│   │   └── task-store.ts         # File-based task state (create, update, complete)
+│   ├── tools/
+│   │   ├── task-plan.ts          # task_plan tool handler
+│   │   ├── task-step-done.ts     # task_step_done tool handler
+│   │   └── task-complete.ts      # task_complete tool handler
+│   └── utils/
+│       └── format.ts             # Progress bar, elapsed time, outcome labels
+├── bootstrap/
+│   └── TASK_TRACKER.md           # Agent behavior instructions
+├── openclaw.plugin.json          # Plugin manifest
+└── openclaw.d.ts                 # Type stubs for openclaw peer dependency
+```
+
+---
+
+## Roadmap
+
+- [ ] `before_tool_call` hook to auto-detect `sessions_spawn` and link sub-agents to parent tasks
+- [ ] Deviation detection: flag steps where output diverges significantly from expected
+- [ ] `task_stuck` tool: pause execution and ask the user for guidance mid-task
+- [ ] Task history CLI: `openclaw doing list` to review past task records
+- [ ] Integration with `OpenClawTaskStatusV1` (openclaw/openclaw#42385) when merged
+
+---
+
+## Contributing
+
+Pull requests welcome. Please open an issue first for significant changes. Repository: [github.com/seekcontext/claw-doing](https://github.com/seekcontext/claw-doing)
+
+## License
+
+MIT

package/SKILL.md

@@ -0,0 +1,182 @@
+---
+name: claw-doing
+description: "Enables real-time progress tracking for long-running tasks in OpenClaw. Use when starting any complex task with 3+ steps, when the user asks for a progress update during long work, or when you are about to do multi-step research, file generation, or sub-agent coordination. Requires the claw-doing plugin: openclaw plugins install claw-doing"
+---
+
+# Claw Doing — Task Progress Tracking
+
+This skill teaches you to use the **claw-doing** plugin to keep users informed during any long or complex task. Instead of going silent while you work, you report progress after every step — directly in the user's chat channel.
+
+## Prerequisites
+
+The `claw-doing` plugin must be installed on the user's OpenClaw gateway:
+
+```bash
+openclaw plugins install claw-doing
+```
+
+After installation, restart the gateway. The plugin registers three tools: `task_plan`, `task_step_done`, and `task_complete`.
+
+## When to use this skill
+
+Use progress tracking whenever the task you are about to perform:
+
+- Has **3 or more distinct steps**
+- Will take **more than ~2 minutes**
+- Involves web research, file generation, data processing, or spawning sub-agents
+- Is the kind of task where the user might wonder "is it still running?"
+
+If in doubt, use it. It costs nothing and significantly improves the user experience.
+
+## Workflow
+
+### Step 1 — Declare the plan (before you do anything else)
+
+The very first thing you do, before executing any step, is call `task_plan`. This registers the task and shows the user what to expect.
+
+```
+task_plan({
+  title: "Research AI agent frameworks",
+  steps: [
+    { index: 1, description: "Search and collect sources", estimatedMinutes: 3 },
+    { index: 2, description: "Extract key findings", estimatedMinutes: 2 },
+    { index: 3, description: "Cross-reference and validate", estimatedMinutes: 2 },
+    { index: 4, description: "Write and deliver report", estimatedMinutes: 3 }
+  ]
+})
+```
+
+The tool returns a formatted plan and a **task ID**. Send the tool output to the user immediately — do not paraphrase or delay it.
+
+Save the task ID: you need it for every subsequent call.
+
+### Step 2 — Report after each step
+
+Immediately after completing a step (before starting the next one), call `task_step_done`:
+
+```
+task_step_done({
+  taskId: "<taskId from task_plan>",
+  stepIndex: 1,
+  summary: "Found 47 sources, filtered to 14 high-quality references",
+  deviation: "Two sources were paywalled — used abstracts only"
+})
+```
+
+The `deviation` field is optional. Use it when something unexpected happened — a source was unavailable, a result was ambiguous, you made a judgment call. This keeps the user informed about decisions you made on their behalf.
+
+Send the tool output to the user immediately.
+
+### Step 3 — Close the task when finished
+
+When the entire task is complete — or if it has definitively failed — call `task_complete`:
+
+```
+task_complete({
+  taskId: "<taskId>",
+  outcome: "success",
+  summary: "Compiled a 2,800-word report covering 8 frameworks with 12 citations. Key finding: LangGraph leads on orchestration; OpenClaw leads on channel integrations.",
+  artifacts: ["memory/2026-03-29-ai-agent-frameworks.md"]
+})
+```
+
+Always call `task_complete`, even on failure — it closes the task record cleanly.
+
+Send the tool output to the user. This is the user's final delivery notification.
+
+## Outcome values
+
+| Value | When to use |
+|-------|-------------|
+| `success` | All steps completed, output matches the user's intent |
+| `partial` | Some steps had issues, but you produced useful output |
+| `failed` | The task could not be completed |
+
+## Rules
+
+- Call `task_plan` **before** executing anything — never retroactively.
+- Call `task_complete` **always** — even when the task fails.
+- Relay tool output to the user **immediately** after each call.
+- Keep `summary` fields **factual and brief** — 1 to 2 sentences.
+- Use `deviation` **honestly** — it builds user trust.
+- Never skip `task_step_done` for a completed step, even if you are in a hurry.
+
+## Full example
+
+```
+User: "Scrape the top 20 Hacker News posts this week and summarize them"
+
+You call:
+  task_plan({
+    title: "Summarize top 20 HN posts",
+    steps: [
+      { index: 1, description: "Fetch HN front page and extract post URLs", estimatedMinutes: 1 },
+      { index: 2, description: "Scrape content from each post", estimatedMinutes: 4 },
+      { index: 3, description: "Generate per-post summaries", estimatedMinutes: 3 },
+      { index: 4, description: "Compile and deliver digest", estimatedMinutes: 2 }
+    ]
+  })
+→ relay output to user
+
+(execute step 1)
+
+  task_step_done({ taskId, stepIndex: 1, summary: "Fetched 20 post URLs from HN front page" })
+→ relay output to user
+
+(execute step 2)
+
+  task_step_done({
+    taskId,
+    stepIndex: 2,
+    summary: "Scraped 18 of 20 posts",
+    deviation: "2 posts linked to PDFs — skipped, will note in digest"
+  })
+→ relay output to user
+
+(execute steps 3–4 similarly)
+
+  task_complete({
+    taskId,
+    outcome: "partial",
+    summary: "Digest covers 18 of 20 posts. Two PDF links were skipped and noted.",
+    artifacts: ["memory/2026-03-29-hn-digest.md"]
+  })
+→ relay output to user
+```
+
+## What the user sees
+
+```
+[░░░░░░░░░░] 0/4 — Starting now...
+
+[███░░░░░░░] 1/4 — elapsed: 1m 2s
+✓ Step 1: Fetched 20 post URLs from HN front page
+
+[██████░░░░] 2/4 — elapsed: 5m 14s
+✓ Step 2: Scraped 18 of 20 posts
+⚠ Note: 2 posts linked to PDFs — skipped, will note in digest
+
+...
+
+[██████████] 4/4 — 10m 33s total
+⚠️ Partially completed
+
+Digest covers 18 of 20 posts. Two PDF links were skipped and noted.
+
+Deliverables:
+  • memory/2026-03-29-hn-digest.md
+```
+
+## FAQ
+
+**What if the task only has 2 steps?**
+Skip the tracking tools — they add overhead without much benefit for very short tasks. Use your judgment: if the user will wait more than 2 minutes, use tracking.
+
+**What if a step fails mid-task?**
+Call `task_step_done` with `deviation` explaining what went wrong, then decide whether to continue, adapt, or call `task_complete` with `outcome: "failed"`.
+
+**What if I don't know the number of steps upfront?**
+Estimate. A rough plan with approximate steps is much better than no plan. You can add a note in `summary` at completion if the actual steps differed.
+
+**Where is the task data stored?**
+In `~/.openclaw/workspace/claw-doing/tasks/<taskId>.json`. Each task is a JSON file the user can inspect or archive.

package/bootstrap/TASK_TRACKER.md

@@ -0,0 +1,78 @@
+# Task Tracker Protocol
+
+You have access to real-time progress tracking tools from the **claw-doing** plugin.
+Use them to keep the user informed during any long or complex task.
+
+## When to use
+
+Use the task tracking tools whenever you are about to perform a task that:
+
+- Has **3 or more distinct steps**
+- Will take **more than ~2 minutes** to complete
+- Involves sub-agent spawning, file generation, web research, or multi-stage processing
+
+## Required workflow
+
+### 1. Declare the plan at the start
+
+Before executing anything, call `task_plan`:
+
+```
+task_plan({
+  title: "Research AI agent frameworks",
+  steps: [
+    { index: 1, description: "Search and collect sources", estimatedMinutes: 3 },
+    { index: 2, description: "Extract key findings", estimatedMinutes: 2 },
+    { index: 3, description: "Cross-reference and validate", estimatedMinutes: 2 },
+    { index: 4, description: "Write and deliver report", estimatedMinutes: 3 }
+  ]
+})
+```
+
+**Immediately relay the tool output to the user** as your message. Do not add extra preamble.
+
+### 2. Report after each step
+
+After you finish each step, call `task_step_done` before starting the next:
+
+```
+task_step_done({
+  taskId: "<taskId from task_plan>",
+  stepIndex: 1,
+  summary: "Found 47 sources, filtered to 14 high-quality references",
+  deviation: "Two sources were paywalled — used abstracts only"  // optional
+})
+```
+
+**Relay the output to the user immediately.** The deviation field is optional — only include it if something unexpected happened.
+
+### 3. Close the task when finished
+
+When the task is fully complete (or has definitively failed), call `task_complete`:
+
+```
+task_complete({
+  taskId: "<taskId>",
+  outcome: "success",  // or "partial" or "failed"
+  summary: "Compiled a 2,800-word report covering 8 frameworks with 12 citations.",
+  artifacts: ["memory/2026-03-29-ai-agent-report.md"]
+})
+```
+
+**Relay the output to the user.** This closes the task record.
+
+## Outcome values
+
+| Outcome | When to use |
+|---------|-------------|
+| `success` | All steps completed, output matches intent |
+| `partial` | Some steps had issues but useful output was produced |
+| `failed` | Task could not be completed — still call this to close the record |
+
+## Rules
+
+- **Always** call `task_plan` before starting a qualifying task — not after.
+- **Always** call `task_complete` at the end, even on failure.
+- Relay tool output directly to the user; do not paraphrase or delay it.
+- Use the `deviation` field honestly — it helps the user understand what changed.
+- Keep step summaries factual and brief (1–2 sentences max).

package/dist/index.d.ts

@@ -0,0 +1,3 @@
+declare const _default: import("openclaw/plugin-sdk/plugin-entry").PluginDefinition;
+export default _default;
+//# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../index.ts"],"names":[],"mappings":";AAaA,wBA2CG"}

package/dist/index.js

@@ -0,0 +1,42 @@
+import { definePluginEntry } from "openclaw/plugin-sdk/plugin-entry";
+import { TaskPlanSchema, executeTaskPlan } from "./src/tools/task-plan.js";
+import { TaskStepDoneSchema, executeTaskStepDone, } from "./src/tools/task-step-done.js";
+import { TaskCompleteSchema, executeTaskComplete, } from "./src/tools/task-complete.js";
+export default definePluginEntry({
+    id: "claw-doing",
+    name: "Claw Doing",
+    description: "Real-time progress tracking for long-running agent tasks. See what your agent is doing — right now.",
+    register(api) {
+        api.registerTool({
+            name: "task_plan",
+            description: "Call this at the START of any complex task that has 3 or more distinct steps. " +
+                "Declares the task plan to the user and returns a task ID needed for subsequent progress calls. " +
+                "Always call task_step_done after completing each step, and task_complete when finished.",
+            parameters: TaskPlanSchema,
+            async execute(_id, params) {
+                return executeTaskPlan(_id, params);
+            },
+        });
+        api.registerTool({
+            name: "task_step_done",
+            description: "Call this immediately after completing each step of a task registered with task_plan. " +
+                "Reports progress to the user with a visual progress bar and step summary. " +
+                "Include a deviation note if the step produced unexpected results.",
+            parameters: TaskStepDoneSchema,
+            async execute(_id, params) {
+                return executeTaskStepDone(_id, params);
+            },
+        });
+        api.registerTool({
+            name: "task_complete",
+            description: "Call this when a task registered with task_plan has finished (success, partial, or failed). " +
+                "Delivers the final summary and list of produced artifacts to the user. " +
+                "Always call this even if the task failed — it closes the task record.",
+            parameters: TaskCompleteSchema,
+            async execute(_id, params) {
+                return executeTaskComplete(_id, params);
+            },
+        });
+    },
+});
+//# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../index.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,iBAAiB,EAAE,MAAM,kCAAkC,CAAC;AACrE,OAAO,EAAE,cAAc,EAAE,eAAe,EAAuB,MAAM,0BAA0B,CAAC;AAChG,OAAO,EACL,kBAAkB,EAClB,mBAAmB,GAEpB,MAAM,+BAA+B,CAAC;AACvC,OAAO,EACL,kBAAkB,EAClB,mBAAmB,GAEpB,MAAM,8BAA8B,CAAC;AAEtC,eAAe,iBAAiB,CAAC;IAC/B,EAAE,EAAE,YAAY;IAChB,IAAI,EAAE,YAAY;IAClB,WAAW,EACT,qGAAqG;IAEvG,QAAQ,CAAC,GAAG;QACV,GAAG,CAAC,YAAY,CAAC;YACf,IAAI,EAAE,WAAW;YACjB,WAAW,EACT,gFAAgF;gBAChF,iGAAiG;gBACjG,yFAAyF;YAC3F,UAAU,EAAE,cAAc;YAC1B,KAAK,CAAC,OAAO,CAAC,GAAG,EAAE,MAAM;gBACvB,OAAO,eAAe,CAAC,GAAG,EAAE,MAAwB,CAAC,CAAC;YACxD,CAAC;SACF,CAAC,CAAC;QAEH,GAAG,CAAC,YAAY,CAAC;YACf,IAAI,EAAE,gBAAgB;YACtB,WAAW,EACT,wFAAwF;gBACxF,4EAA4E;gBAC5E,mEAAmE;YACrE,UAAU,EAAE,kBAAkB;YAC9B,KAAK,CAAC,OAAO,CAAC,GAAG,EAAE,MAAM;gBACvB,OAAO,mBAAmB,CAAC,GAAG,EAAE,MAA4B,CAAC,CAAC;YAChE,CAAC;SACF,CAAC,CAAC;QAEH,GAAG,CAAC,YAAY,CAAC;YACf,IAAI,EAAE,eAAe;YACrB,WAAW,EACT,8FAA8F;gBAC9F,yEAAyE;gBACzE,uEAAuE;YACzE,UAAU,EAAE,kBAAkB;YAC9B,KAAK,CAAC,OAAO,CAAC,GAAG,EAAE,MAAM;gBACvB,OAAO,mBAAmB,CAAC,GAAG,EAAE,MAA4B,CAAC,CAAC;YAChE,CAAC;SACF,CAAC,CAAC;IACL,CAAC;CACF,CAAC,CAAC"}

package/dist/src/store/task-store.d.ts

@@ -0,0 +1,11 @@
+import type { Task } from "../types.js";
+export declare function saveTask(task: Task): Promise<void>;
+export declare function loadTask(taskId: string): Promise<Task | null>;
+export declare function createTask(title: string, rawSteps: Array<{
+    index: number;
+    description: string;
+    estimatedMinutes?: number;
+}>): Task;
+export declare function markStepDone(taskId: string, stepIndex: number, summary: string, deviation?: string): Promise<Task | null>;
+export declare function completeTask(taskId: string, outcome: "success" | "partial" | "failed", summary: string, artifacts?: string[]): Promise<Task | null>;
+//# sourceMappingURL=task-store.d.ts.map

package/dist/src/store/task-store.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"task-store.d.ts","sourceRoot":"","sources":["../../../src/store/task-store.ts"],"names":[],"mappings":"AAGA,OAAO,KAAK,EAAE,IAAI,EAAY,MAAM,aAAa,CAAC;AAiBlD,wBAAsB,QAAQ,CAAC,IAAI,EAAE,IAAI,GAAG,OAAO,CAAC,IAAI,CAAC,CAIxD;AAED,wBAAsB,QAAQ,CAAC,MAAM,EAAE,MAAM,GAAG,OAAO,CAAC,IAAI,GAAG,IAAI,CAAC,CAOnE;AAED,wBAAgB,UAAU,CACxB,KAAK,EAAE,MAAM,EACb,QAAQ,EAAE,KAAK,CAAC;IAAE,KAAK,EAAE,MAAM,CAAC;IAAC,WAAW,EAAE,MAAM,CAAC;IAAC,gBAAgB,CAAC,EAAE,MAAM,CAAA;CAAE,CAAC,GACjF,IAAI,CAmBN;AAED,wBAAsB,YAAY,CAChC,MAAM,EAAE,MAAM,EACd,SAAS,EAAE,MAAM,EACjB,OAAO,EAAE,MAAM,EACf,SAAS,CAAC,EAAE,MAAM,GACjB,OAAO,CAAC,IAAI,GAAG,IAAI,CAAC,CAiBtB;AAED,wBAAsB,YAAY,CAChC,MAAM,EAAE,MAAM,EACd,OAAO,EAAE,SAAS,GAAG,SAAS,GAAG,QAAQ,EACzC,OAAO,EAAE,MAAM,EACf,SAAS,CAAC,EAAE,MAAM,EAAE,GACnB,OAAO,CAAC,IAAI,GAAG,IAAI,CAAC,CAmBtB"}

package/dist/src/store/task-store.js

@@ -0,0 +1,81 @@
+import { promises as fs } from "node:fs";
+import { join } from "node:path";
+import { homedir } from "node:os";
+function getStorageDir() {
+    return (process.env["CLAW_DOING_STORAGE_DIR"] ??
+        join(homedir(), ".openclaw", "workspace", "claw-doing", "tasks"));
+}
+async function ensureDir(dir) {
+    await fs.mkdir(dir, { recursive: true });
+}
+function taskFilePath(taskId) {
+    return join(getStorageDir(), `${taskId}.json`);
+}
+export async function saveTask(task) {
+    const dir = getStorageDir();
+    await ensureDir(dir);
+    await fs.writeFile(taskFilePath(task.taskId), JSON.stringify(task, null, 2), "utf-8");
+}
+export async function loadTask(taskId) {
+    try {
+        const raw = await fs.readFile(taskFilePath(taskId), "utf-8");
+        return JSON.parse(raw);
+    }
+    catch {
+        return null;
+    }
+}
+export function createTask(title, rawSteps) {
+    const now = new Date().toISOString();
+    const steps = rawSteps.map((s) => ({
+        index: s.index,
+        description: s.description,
+        estimatedMinutes: s.estimatedMinutes,
+        status: "pending",
+    }));
+    return {
+        taskId: crypto.randomUUID(),
+        title,
+        steps,
+        totalSteps: steps.length,
+        completedSteps: 0,
+        status: "running",
+        createdAt: now,
+        updatedAt: now,
+    };
+}
+export async function markStepDone(taskId, stepIndex, summary, deviation) {
+    const task = await loadTask(taskId);
+    if (!task)
+        return null;
+    const step = task.steps.find((s) => s.index === stepIndex);
+    if (!step)
+        return null;
+    step.status = "done";
+    step.summary = summary;
+    step.deviation = deviation;
+    step.completedAt = new Date().toISOString();
+    task.completedSteps = task.steps.filter((s) => s.status === "done").length;
+    task.updatedAt = new Date().toISOString();
+    await saveTask(task);
+    return task;
+}
+export async function completeTask(taskId, outcome, summary, artifacts) {
+    const task = await loadTask(taskId);
+    if (!task)
+        return null;
+    const statusMap = {
+        success: "done",
+        partial: "partial",
+        failed: "failed",
+    };
+    task.status = statusMap[outcome];
+    task.outcome = outcome;
+    task.finalSummary = summary;
+    task.artifacts = artifacts;
+    task.completedAt = new Date().toISOString();
+    task.updatedAt = new Date().toISOString();
+    await saveTask(task);
+    return task;
+}
+//# sourceMappingURL=task-store.js.map

package/dist/src/store/task-store.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"task-store.js","sourceRoot":"","sources":["../../../src/store/task-store.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,QAAQ,IAAI,EAAE,EAAE,MAAM,SAAS,CAAC;AACzC,OAAO,EAAE,IAAI,EAAE,MAAM,WAAW,CAAC;AACjC,OAAO,EAAE,OAAO,EAAE,MAAM,SAAS,CAAC;AAGlC,SAAS,aAAa;IACpB,OAAO,CACL,OAAO,CAAC,GAAG,CAAC,wBAAwB,CAAC;QACrC,IAAI,CAAC,OAAO,EAAE,EAAE,WAAW,EAAE,WAAW,EAAE,YAAY,EAAE,OAAO,CAAC,CACjE,CAAC;AACJ,CAAC;AAED,KAAK,UAAU,SAAS,CAAC,GAAW;IAClC,MAAM,EAAE,CAAC,KAAK,CAAC,GAAG,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAC;AAC3C,CAAC;AAED,SAAS,YAAY,CAAC,MAAc;IAClC,OAAO,IAAI,CAAC,aAAa,EAAE,EAAE,GAAG,MAAM,OAAO,CAAC,CAAC;AACjD,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,QAAQ,CAAC,IAAU;IACvC,MAAM,GAAG,GAAG,aAAa,EAAE,CAAC;IAC5B,MAAM,SAAS,CAAC,GAAG,CAAC,CAAC;IACrB,MAAM,EAAE,CAAC,SAAS,CAAC,YAAY,CAAC,IAAI,CAAC,MAAM,CAAC,EAAE,IAAI,CAAC,SAAS,CAAC,IAAI,EAAE,IAAI,EAAE,CAAC,CAAC,EAAE,OAAO,CAAC,CAAC;AACxF,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,QAAQ,CAAC,MAAc;IAC3C,IAAI,CAAC;QACH,MAAM,GAAG,GAAG,MAAM,EAAE,CAAC,QAAQ,CAAC,YAAY,CAAC,MAAM,CAAC,EAAE,OAAO,CAAC,CAAC;QAC7D,OAAO,IAAI,CAAC,KAAK,CAAC,GAAG,CAAS,CAAC;IACjC,CAAC;IAAC,MAAM,CAAC;QACP,OAAO,IAAI,CAAC;IACd,CAAC;AACH,CAAC;AAED,MAAM,UAAU,UAAU,CACxB,KAAa,EACb,QAAkF;IAElF,MAAM,GAAG,GAAG,IAAI,IAAI,EAAE,CAAC,WAAW,EAAE,CAAC;IACrC,MAAM,KAAK,GAAe,QAAQ,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC;QAC7C,KAAK,EAAE,CAAC,CAAC,KAAK;QACd,WAAW,EAAE,CAAC,CAAC,WAAW;QAC1B,gBAAgB,EAAE,CAAC,CAAC,gBAAgB;QACpC,MAAM,EAAE,SAAS;KAClB,CAAC,CAAC,CAAC;IAEJ,OAAO;QACL,MAAM,EAAE,MAAM,CAAC,UAAU,EAAE;QAC3B,KAAK;QACL,KAAK;QACL,UAAU,EAAE,KAAK,CAAC,MAAM;QACxB,cAAc,EAAE,CAAC;QACjB,MAAM,EAAE,SAAS;QACjB,SAAS,EAAE,GAAG;QACd,SAAS,EAAE,GAAG;KACf,CAAC;AACJ,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,YAAY,CAChC,MAAc,EACd,SAAiB,EACjB,OAAe,EACf,SAAkB;IAElB,MAAM,IAAI,GAAG,MAAM,QAAQ,CAAC,MAAM,CAAC,CAAC;IACpC,IAAI,CAAC,IAAI;QAAE,OAAO,IAAI,CAAC;IAEvB,MAAM,IAAI,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,KAAK,KAAK,SAAS,CAAC,CAAC;IAC3D,IAAI,CAAC,IAAI;QAAE,OAAO,IAAI,CAAC;IAEvB,IAAI,CAAC,MAAM,GAAG,MAAM,CAAC;IACrB,IAAI,CAAC,OAAO,GAAG,OAAO,CAAC;IACvB,IAAI,CAAC,SAAS,GAAG,SAAS,CAAC;IAC3B,IAAI,CAAC,WAAW,GAAG,IAAI,IAAI,EAAE,CAAC,WAAW,EAAE,CAAC;IAE5C,IAAI,CAAC,cAAc,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,KAAK,MAAM,CAAC,CAAC,MAAM,CAAC;IAC3E,IAAI,CAAC,SAAS,GAAG,IAAI,IAAI,EAAE,CAAC,WAAW,EAAE,CAAC;IAE1C,MAAM,QAAQ,CAAC,IAAI,CAAC,CAAC;IACrB,OAAO,IAAI,CAAC;AACd,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,YAAY,CAChC,MAAc,EACd,OAAyC,EACzC,OAAe,EACf,SAAoB;IAEpB,MAAM,IAAI,GAAG,MAAM,QAAQ,CAAC,MAAM,CAAC,CAAC;IACpC,IAAI,CAAC,IAAI;QAAE,OAAO,IAAI,CAAC;IAEvB,MAAM,SAAS,GAA2C;QACxD,OAAO,EAAE,MAAM;QACf,OAAO,EAAE,SAAS;QAClB,MAAM,EAAE,QAAQ;KACjB,CAAC;IAEF,IAAI,CAAC,MAAM,GAAG,SAAS,CAAC,OAAO,CAAC,CAAC;IACjC,IAAI,CAAC,OAAO,GAAG,OAAO,CAAC;IACvB,IAAI,CAAC,YAAY,GAAG,OAAO,CAAC;IAC5B,IAAI,CAAC,SAAS,GAAG,SAAS,CAAC;IAC3B,IAAI,CAAC,WAAW,GAAG,IAAI,IAAI,EAAE,CAAC,WAAW,EAAE,CAAC;IAC5C,IAAI,CAAC,SAAS,GAAG,IAAI,IAAI,EAAE,CAAC,WAAW,EAAE,CAAC;IAE1C,MAAM,QAAQ,CAAC,IAAI,CAAC,CAAC;IACrB,OAAO,IAAI,CAAC;AACd,CAAC"}

package/dist/src/tools/task-complete.d.ts

@@ -0,0 +1,19 @@
+export declare const TaskCompleteSchema: import("@sinclair/typebox").TObject<{
+    taskId: import("@sinclair/typebox").TString;
+    outcome: import("@sinclair/typebox").TUnion<[import("@sinclair/typebox").TLiteral<"success">, import("@sinclair/typebox").TLiteral<"partial">, import("@sinclair/typebox").TLiteral<"failed">]>;
+    summary: import("@sinclair/typebox").TString;
+    artifacts: import("@sinclair/typebox").TOptional<import("@sinclair/typebox").TArray<import("@sinclair/typebox").TString>>;
+}>;
+export type TaskCompleteParams = {
+    taskId: string;
+    outcome: "success" | "partial" | "failed";
+    summary: string;
+    artifacts?: string[];
+};
+export declare function executeTaskComplete(_id: string, params: TaskCompleteParams): Promise<{
+    content: {
+        type: "text";
+        text: string;
+    }[];
+}>;
+//# sourceMappingURL=task-complete.d.ts.map

package/dist/src/tools/task-complete.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"task-complete.d.ts","sourceRoot":"","sources":["../../../src/tools/task-complete.ts"],"names":[],"mappings":"AAIA,eAAO,MAAM,kBAAkB;;;;;EAmB7B,CAAC;AAEH,MAAM,MAAM,kBAAkB,GAAG;IAC/B,MAAM,EAAE,MAAM,CAAC;IACf,OAAO,EAAE,SAAS,GAAG,SAAS,GAAG,QAAQ,CAAC;IAC1C,OAAO,EAAE,MAAM,CAAC;IAChB,SAAS,CAAC,EAAE,MAAM,EAAE,CAAC;CACtB,CAAC;AAEF,wBAAsB,mBAAmB,CAAC,GAAG,EAAE,MAAM,EAAE,MAAM,EAAE,kBAAkB;;;;;GA2ChF"}

package/dist/src/tools/task-complete.js

@@ -0,0 +1,49 @@
+import { Type } from "@sinclair/typebox";
+import { completeTask } from "../store/task-store.js";
+import { buildProgressBar, formatElapsed, formatOutcome } from "../utils/format.js";
+export const TaskCompleteSchema = Type.Object({
+    taskId: Type.String({ description: "The task ID returned by task_plan" }),
+    outcome: Type.Union([Type.Literal("success"), Type.Literal("partial"), Type.Literal("failed")], {
+        description: "'success' if all steps completed as planned, 'partial' if some steps had issues but useful output was produced, 'failed' if the task could not be completed",
+    }),
+    summary: Type.String({
+        description: "A clear, user-facing summary of what was accomplished. Include key numbers, findings, or decisions made.",
+    }),
+    artifacts: Type.Optional(Type.Array(Type.String(), {
+        description: "List of output file paths, URLs, or named results produced by this task (e.g. 'memory/2026-03-29-report.md')",
+    })),
+});
+export async function executeTaskComplete(_id, params) {
+    const task = await completeTask(params.taskId, params.outcome, params.summary, params.artifacts);
+    if (!task) {
+        return {
+            content: [
+                {
+                    type: "text",
+                    text: `Error: Task \`${params.taskId}\` not found. Please check the task ID.`,
+                },
+            ],
+        };
+    }
+    const bar = buildProgressBar(task.completedSteps, task.totalSteps);
+    const elapsed = formatElapsed(task.createdAt);
+    const outcomeLabel = formatOutcome(params.outcome);
+    const lines = [
+        `${bar} — ${elapsed} total`,
+        ``,
+        `${outcomeLabel}`,
+        ``,
+        params.summary,
+    ];
+    if (params.artifacts && params.artifacts.length > 0) {
+        lines.push(``, `Deliverables:`);
+        params.artifacts.forEach((a) => lines.push(`  • ${a}`));
+    }
+    const failedSteps = task.steps.filter((s) => s.status === "failed");
+    if (failedSteps.length > 0) {
+        lines.push(``, `Incomplete steps:`);
+        failedSteps.forEach((s) => lines.push(`  • Step ${s.index}: ${s.description}`));
+    }
+    return { content: [{ type: "text", text: lines.join("\n") }] };
+}
+//# sourceMappingURL=task-complete.js.map

package/dist/src/tools/task-complete.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"task-complete.js","sourceRoot":"","sources":["../../../src/tools/task-complete.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,IAAI,EAAE,MAAM,mBAAmB,CAAC;AACzC,OAAO,EAAE,YAAY,EAAE,MAAM,wBAAwB,CAAC;AACtD,OAAO,EAAE,gBAAgB,EAAE,aAAa,EAAE,aAAa,EAAE,MAAM,oBAAoB,CAAC;AAEpF,MAAM,CAAC,MAAM,kBAAkB,GAAG,IAAI,CAAC,MAAM,CAAC;IAC5C,MAAM,EAAE,IAAI,CAAC,MAAM,CAAC,EAAE,WAAW,EAAE,mCAAmC,EAAE,CAAC;IACzE,OAAO,EAAE,IAAI,CAAC,KAAK,CACjB,CAAC,IAAI,CAAC,OAAO,CAAC,SAAS,CAAC,EAAE,IAAI,CAAC,OAAO,CAAC,SAAS,CAAC,EAAE,IAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAC,EAC1E;QACE,WAAW,EACT,6JAA6J;KAChK,CACF;IACD,OAAO,EAAE,IAAI,CAAC,MAAM,CAAC;QACnB,WAAW,EACT,0GAA0G;KAC7G,CAAC;IACF,SAAS,EAAE,IAAI,CAAC,QAAQ,CACtB,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,MAAM,EAAE,EAAE;QACxB,WAAW,EACT,8GAA8G;KACjH,CAAC,CACH;CACF,CAAC,CAAC;AASH,MAAM,CAAC,KAAK,UAAU,mBAAmB,CAAC,GAAW,EAAE,MAA0B;IAC/E,MAAM,IAAI,GAAG,MAAM,YAAY,CAC7B,MAAM,CAAC,MAAM,EACb,MAAM,CAAC,OAAO,EACd,MAAM,CAAC,OAAO,EACd,MAAM,CAAC,SAAS,CACjB,CAAC;IAEF,IAAI,CAAC,IAAI,EAAE,CAAC;QACV,OAAO;YACL,OAAO,EAAE;gBACP;oBACE,IAAI,EAAE,MAAe;oBACrB,IAAI,EAAE,iBAAiB,MAAM,CAAC,MAAM,yCAAyC;iBAC9E;aACF;SACF,CAAC;IACJ,CAAC;IAED,MAAM,GAAG,GAAG,gBAAgB,CAAC,IAAI,CAAC,cAAc,EAAE,IAAI,CAAC,UAAU,CAAC,CAAC;IACnE,MAAM,OAAO,GAAG,aAAa,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC;IAC9C,MAAM,YAAY,GAAG,aAAa,CAAC,MAAM,CAAC,OAAO,CAAC,CAAC;IAEnD,MAAM,KAAK,GAAG;QACZ,GAAG,GAAG,MAAM,OAAO,QAAQ;QAC3B,EAAE;QACF,GAAG,YAAY,EAAE;QACjB,EAAE;QACF,MAAM,CAAC,OAAO;KACf,CAAC;IAEF,IAAI,MAAM,CAAC,SAAS,IAAI,MAAM,CAAC,SAAS,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QACpD,KAAK,CAAC,IAAI,CAAC,EAAE,EAAE,eAAe,CAAC,CAAC;QAChC,MAAM,CAAC,SAAS,CAAC,OAAO,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,OAAO,CAAC,EAAE,CAAC,CAAC,CAAC;IAC1D,CAAC;IAED,MAAM,WAAW,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,KAAK,QAAQ,CAAC,CAAC;IACpE,IAAI,WAAW,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QAC3B,KAAK,CAAC,IAAI,CAAC,EAAE,EAAE,mBAAmB,CAAC,CAAC;QACpC,WAAW,CAAC,OAAO,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,YAAY,CAAC,CAAC,KAAK,KAAK,CAAC,CAAC,WAAW,EAAE,CAAC,CAAC,CAAC;IAClF,CAAC;IAED,OAAO,EAAE,OAAO,EAAE,CAAC,EAAE,IAAI,EAAE,MAAe,EAAE,IAAI,EAAE,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC,EAAE,CAAC;AAC1E,CAAC"}

package/dist/src/tools/task-plan.d.ts

@@ -0,0 +1,23 @@
+export declare const TaskPlanSchema: import("@sinclair/typebox").TObject<{
+    title: import("@sinclair/typebox").TString;
+    steps: import("@sinclair/typebox").TArray<import("@sinclair/typebox").TObject<{
+        index: import("@sinclair/typebox").TNumber;
+        description: import("@sinclair/typebox").TString;
+        estimatedMinutes: import("@sinclair/typebox").TOptional<import("@sinclair/typebox").TNumber>;
+    }>>;
+}>;
+export type TaskPlanParams = {
+    title: string;
+    steps: Array<{
+        index: number;
+        description: string;
+        estimatedMinutes?: number;
+    }>;
+};
+export declare function executeTaskPlan(_id: string, params: TaskPlanParams): Promise<{
+    content: {
+        type: "text";
+        text: string;
+    }[];
+}>;
+//# sourceMappingURL=task-plan.d.ts.map

package/dist/src/tools/task-plan.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"task-plan.d.ts","sourceRoot":"","sources":["../../../src/tools/task-plan.ts"],"names":[],"mappings":"AAIA,eAAO,MAAM,cAAc;;;;;;;EAczB,CAAC;AAEH,MAAM,MAAM,cAAc,GAAG;IAC3B,KAAK,EAAE,MAAM,CAAC;IACd,KAAK,EAAE,KAAK,CAAC;QAAE,KAAK,EAAE,MAAM,CAAC;QAAC,WAAW,EAAE,MAAM,CAAC;QAAC,gBAAgB,CAAC,EAAE,MAAM,CAAA;KAAE,CAAC,CAAC;CACjF,CAAC;AAEF,wBAAsB,eAAe,CAAC,GAAG,EAAE,MAAM,EAAE,MAAM,EAAE,cAAc;;;;;GAyBxE"}

package/dist/src/tools/task-plan.js

@@ -0,0 +1,36 @@
+import { Type } from "@sinclair/typebox";
+import { createTask, saveTask } from "../store/task-store.js";
+import { buildProgressBar, formatDuration } from "../utils/format.js";
+export const TaskPlanSchema = Type.Object({
+    title: Type.String({
+        description: "A short title describing the overall task (e.g. 'Research AI agent frameworks')",
+    }),
+    steps: Type.Array(Type.Object({
+        index: Type.Number({ description: "Step number, starting from 1" }),
+        description: Type.String({ description: "What this step does" }),
+        estimatedMinutes: Type.Optional(Type.Number({ description: "Estimated duration in minutes" })),
+    }), { minItems: 1, description: "Ordered list of steps to complete the task" }),
+});
+export async function executeTaskPlan(_id, params) {
+    const task = createTask(params.title, params.steps);
+    await saveTask(task);
+    const stepList = params.steps
+        .map((s) => {
+        const eta = s.estimatedMinutes ? ` (${formatDuration(s.estimatedMinutes)})` : "";
+        return `  ${s.index}. ${s.description}${eta}`;
+    })
+        .join("\n");
+    const bar = buildProgressBar(0, params.steps.length);
+    const message = [
+        `**${params.title}**`,
+        `Task ID: \`${task.taskId}\``,
+        ``,
+        `Plan — ${params.steps.length} step${params.steps.length > 1 ? "s" : ""}:`,
+        stepList,
+        ``,
+        `${bar}`,
+        `Starting now...`,
+    ].join("\n");
+    return { content: [{ type: "text", text: message }] };
+}
+//# sourceMappingURL=task-plan.js.map

package/dist/src/tools/task-plan.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"task-plan.js","sourceRoot":"","sources":["../../../src/tools/task-plan.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,IAAI,EAAE,MAAM,mBAAmB,CAAC;AACzC,OAAO,EAAE,UAAU,EAAE,QAAQ,EAAE,MAAM,wBAAwB,CAAC;AAC9D,OAAO,EAAE,gBAAgB,EAAE,cAAc,EAAE,MAAM,oBAAoB,CAAC;AAEtE,MAAM,CAAC,MAAM,cAAc,GAAG,IAAI,CAAC,MAAM,CAAC;IACxC,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC;QACjB,WAAW,EAAE,iFAAiF;KAC/F,CAAC;IACF,KAAK,EAAE,IAAI,CAAC,KAAK,CACf,IAAI,CAAC,MAAM,CAAC;QACV,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,EAAE,WAAW,EAAE,8BAA8B,EAAE,CAAC;QACnE,WAAW,EAAE,IAAI,CAAC,MAAM,CAAC,EAAE,WAAW,EAAE,qBAAqB,EAAE,CAAC;QAChE,gBAAgB,EAAE,IAAI,CAAC,QAAQ,CAC7B,IAAI,CAAC,MAAM,CAAC,EAAE,WAAW,EAAE,+BAA+B,EAAE,CAAC,CAC9D;KACF,CAAC,EACF,EAAE,QAAQ,EAAE,CAAC,EAAE,WAAW,EAAE,4CAA4C,EAAE,CAC3E;CACF,CAAC,CAAC;AAOH,MAAM,CAAC,KAAK,UAAU,eAAe,CAAC,GAAW,EAAE,MAAsB;IACvE,MAAM,IAAI,GAAG,UAAU,CAAC,MAAM,CAAC,KAAK,EAAE,MAAM,CAAC,KAAK,CAAC,CAAC;IACpD,MAAM,QAAQ,CAAC,IAAI,CAAC,CAAC;IAErB,MAAM,QAAQ,GAAG,MAAM,CAAC,KAAK;SAC1B,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE;QACT,MAAM,GAAG,GAAG,CAAC,CAAC,gBAAgB,CAAC,CAAC,CAAC,KAAK,cAAc,CAAC,CAAC,CAAC,gBAAgB,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC;QACjF,OAAO,KAAK,CAAC,CAAC,KAAK,KAAK,CAAC,CAAC,WAAW,GAAG,GAAG,EAAE,CAAC;IAChD,CAAC,CAAC;SACD,IAAI,CAAC,IAAI,CAAC,CAAC;IAEd,MAAM,GAAG,GAAG,gBAAgB,CAAC,CAAC,EAAE,MAAM,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC;IAErD,MAAM,OAAO,GAAG;QACd,KAAK,MAAM,CAAC,KAAK,IAAI;QACrB,cAAc,IAAI,CAAC,MAAM,IAAI;QAC7B,EAAE;QACF,UAAU,MAAM,CAAC,KAAK,CAAC,MAAM,QAAQ,MAAM,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,GAAG;QAC1E,QAAQ;QACR,EAAE;QACF,GAAG,GAAG,EAAE;QACR,iBAAiB;KAClB,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IAEb,OAAO,EAAE,OAAO,EAAE,CAAC,EAAE,IAAI,EAAE,MAAe,EAAE,IAAI,EAAE,OAAO,EAAE,CAAC,EAAE,CAAC;AACjE,CAAC"}

package/dist/src/tools/task-step-done.d.ts

@@ -0,0 +1,19 @@
+export declare const TaskStepDoneSchema: import("@sinclair/typebox").TObject<{
+    taskId: import("@sinclair/typebox").TString;
+    stepIndex: import("@sinclair/typebox").TNumber;
+    summary: import("@sinclair/typebox").TString;
+    deviation: import("@sinclair/typebox").TOptional<import("@sinclair/typebox").TString>;
+}>;
+export type TaskStepDoneParams = {
+    taskId: string;
+    stepIndex: number;
+    summary: string;
+    deviation?: string;
+};
+export declare function executeTaskStepDone(_id: string, params: TaskStepDoneParams): Promise<{
+    content: {
+        type: "text";
+        text: string;
+    }[];
+}>;
+//# sourceMappingURL=task-step-done.d.ts.map

package/dist/src/tools/task-step-done.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"task-step-done.d.ts","sourceRoot":"","sources":["../../../src/tools/task-step-done.ts"],"names":[],"mappings":"AAIA,eAAO,MAAM,kBAAkB;;;;;EAY7B,CAAC;AAEH,MAAM,MAAM,kBAAkB,GAAG;IAC/B,MAAM,EAAE,MAAM,CAAC;IACf,SAAS,EAAE,MAAM,CAAC;IAClB,OAAO,EAAE,MAAM,CAAC;IAChB,SAAS,CAAC,EAAE,MAAM,CAAC;CACpB,CAAC;AAEF,wBAAsB,mBAAmB,CAAC,GAAG,EAAE,MAAM,EAAE,MAAM,EAAE,kBAAkB;;;;;GAyChF"}

package/dist/src/tools/task-step-done.js

@@ -0,0 +1,45 @@
+import { Type } from "@sinclair/typebox";
+import { markStepDone } from "../store/task-store.js";
+import { buildProgressBar, formatElapsed } from "../utils/format.js";
+export const TaskStepDoneSchema = Type.Object({
+    taskId: Type.String({ description: "The task ID returned by task_plan" }),
+    stepIndex: Type.Number({ description: "The step number that was just completed" }),
+    summary: Type.String({
+        description: "A concise description of what was accomplished in this step",
+    }),
+    deviation: Type.Optional(Type.String({
+        description: "Describe any unexpected finding or deviation from the original plan (optional). Leave empty if everything went as expected.",
+    })),
+});
+export async function executeTaskStepDone(_id, params) {
+    const task = await markStepDone(params.taskId, params.stepIndex, params.summary, params.deviation);
+    if (!task) {
+        return {
+            content: [
+                {
+                    type: "text",
+                    text: `Error: Task \`${params.taskId}\` not found. Please check the task ID.`,
+                },
+            ],
+        };
+    }
+    const bar = buildProgressBar(task.completedSteps, task.totalSteps);
+    const elapsed = formatElapsed(task.createdAt);
+    const isLastStep = task.completedSteps === task.totalSteps;
+    const lines = [
+        `${bar} — elapsed: ${elapsed}`,
+        ``,
+        `✓ Step ${params.stepIndex}: ${params.summary}`,
+    ];
+    if (params.deviation) {
+        lines.push(``, `⚠ Note: ${params.deviation}`);
+    }
+    if (!isLastStep) {
+        const nextStep = task.steps.find((s) => s.status === "pending");
+        if (nextStep) {
+            lines.push(``, `Next: Step ${nextStep.index} — ${nextStep.description}`);
+        }
+    }
+    return { content: [{ type: "text", text: lines.join("\n") }] };
+}
+//# sourceMappingURL=task-step-done.js.map

package/dist/src/tools/task-step-done.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"task-step-done.js","sourceRoot":"","sources":["../../../src/tools/task-step-done.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,IAAI,EAAE,MAAM,mBAAmB,CAAC;AACzC,OAAO,EAAE,YAAY,EAAE,MAAM,wBAAwB,CAAC;AACtD,OAAO,EAAE,gBAAgB,EAAE,aAAa,EAAE,MAAM,oBAAoB,CAAC;AAErE,MAAM,CAAC,MAAM,kBAAkB,GAAG,IAAI,CAAC,MAAM,CAAC;IAC5C,MAAM,EAAE,IAAI,CAAC,MAAM,CAAC,EAAE,WAAW,EAAE,mCAAmC,EAAE,CAAC;IACzE,SAAS,EAAE,IAAI,CAAC,MAAM,CAAC,EAAE,WAAW,EAAE,yCAAyC,EAAE,CAAC;IAClF,OAAO,EAAE,IAAI,CAAC,MAAM,CAAC;QACnB,WAAW,EAAE,6DAA6D;KAC3E,CAAC;IACF,SAAS,EAAE,IAAI,CAAC,QAAQ,CACtB,IAAI,CAAC,MAAM,CAAC;QACV,WAAW,EACT,6HAA6H;KAChI,CAAC,CACH;CACF,CAAC,CAAC;AASH,MAAM,CAAC,KAAK,UAAU,mBAAmB,CAAC,GAAW,EAAE,MAA0B;IAC/E,MAAM,IAAI,GAAG,MAAM,YAAY,CAC7B,MAAM,CAAC,MAAM,EACb,MAAM,CAAC,SAAS,EAChB,MAAM,CAAC,OAAO,EACd,MAAM,CAAC,SAAS,CACjB,CAAC;IAEF,IAAI,CAAC,IAAI,EAAE,CAAC;QACV,OAAO;YACL,OAAO,EAAE;gBACP;oBACE,IAAI,EAAE,MAAe;oBACrB,IAAI,EAAE,iBAAiB,MAAM,CAAC,MAAM,yCAAyC;iBAC9E;aACF;SACF,CAAC;IACJ,CAAC;IAED,MAAM,GAAG,GAAG,gBAAgB,CAAC,IAAI,CAAC,cAAc,EAAE,IAAI,CAAC,UAAU,CAAC,CAAC;IACnE,MAAM,OAAO,GAAG,aAAa,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC;IAC9C,MAAM,UAAU,GAAG,IAAI,CAAC,cAAc,KAAK,IAAI,CAAC,UAAU,CAAC;IAE3D,MAAM,KAAK,GAAG;QACZ,GAAG,GAAG,eAAe,OAAO,EAAE;QAC9B,EAAE;QACF,UAAU,MAAM,CAAC,SAAS,KAAK,MAAM,CAAC,OAAO,EAAE;KAChD,CAAC;IAEF,IAAI,MAAM,CAAC,SAAS,EAAE,CAAC;QACrB,KAAK,CAAC,IAAI,CAAC,EAAE,EAAE,WAAW,MAAM,CAAC,SAAS,EAAE,CAAC,CAAC;IAChD,CAAC;IAED,IAAI,CAAC,UAAU,EAAE,CAAC;QAChB,MAAM,QAAQ,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,KAAK,SAAS,CAAC,CAAC;QAChE,IAAI,QAAQ,EAAE,CAAC;YACb,KAAK,CAAC,IAAI,CAAC,EAAE,EAAE,cAAc,QAAQ,CAAC,KAAK,MAAM,QAAQ,CAAC,WAAW,EAAE,CAAC,CAAC;QAC3E,CAAC;IACH,CAAC;IAED,OAAO,EAAE,OAAO,EAAE,CAAC,EAAE,IAAI,EAAE,MAAe,EAAE,IAAI,EAAE,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC,EAAE,CAAC;AAC1E,CAAC"}

package/dist/src/types.d.ts

@@ -0,0 +1,28 @@
+export interface TaskStep {
+    index: number;
+    description: string;
+    estimatedMinutes?: number;
+    status: "pending" | "done" | "failed";
+    summary?: string;
+    deviation?: string;
+    completedAt?: string;
+}
+export interface Task {
+    taskId: string;
+    title: string;
+    steps: TaskStep[];
+    totalSteps: number;
+    completedSteps: number;
+    status: "running" | "done" | "partial" | "failed";
+    createdAt: string;
+    updatedAt: string;
+    completedAt?: string;
+    outcome?: "success" | "partial" | "failed";
+    finalSummary?: string;
+    artifacts?: string[];
+}
+export interface PluginConfig {
+    storageDir?: string;
+    progressBarWidth?: number;
+}
+//# sourceMappingURL=types.d.ts.map

package/dist/src/types.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"types.d.ts","sourceRoot":"","sources":["../../src/types.ts"],"names":[],"mappings":"AAAA,MAAM,WAAW,QAAQ;IACvB,KAAK,EAAE,MAAM,CAAC;IACd,WAAW,EAAE,MAAM,CAAC;IACpB,gBAAgB,CAAC,EAAE,MAAM,CAAC;IAC1B,MAAM,EAAE,SAAS,GAAG,MAAM,GAAG,QAAQ,CAAC;IACtC,OAAO,CAAC,EAAE,MAAM,CAAC;IACjB,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB,WAAW,CAAC,EAAE,MAAM,CAAC;CACtB;AAED,MAAM,WAAW,IAAI;IACnB,MAAM,EAAE,MAAM,CAAC;IACf,KAAK,EAAE,MAAM,CAAC;IACd,KAAK,EAAE,QAAQ,EAAE,CAAC;IAClB,UAAU,EAAE,MAAM,CAAC;IACnB,cAAc,EAAE,MAAM,CAAC;IACvB,MAAM,EAAE,SAAS,GAAG,MAAM,GAAG,SAAS,GAAG,QAAQ,CAAC;IAClD,SAAS,EAAE,MAAM,CAAC;IAClB,SAAS,EAAE,MAAM,CAAC;IAClB,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,OAAO,CAAC,EAAE,SAAS,GAAG,SAAS,GAAG,QAAQ,CAAC;IAC3C,YAAY,CAAC,EAAE,MAAM,CAAC;IACtB,SAAS,CAAC,EAAE,MAAM,EAAE,CAAC;CACtB;AAED,MAAM,WAAW,YAAY;IAC3B,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB,gBAAgB,CAAC,EAAE,MAAM,CAAC;CAC3B"}

package/dist/src/types.js

@@ -0,0 +1,2 @@
+export {};
+//# sourceMappingURL=types.js.map

package/dist/src/types.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"types.js","sourceRoot":"","sources":["../../src/types.ts"],"names":[],"mappings":""}

package/dist/src/utils/format.d.ts

@@ -0,0 +1,5 @@
+export declare function buildProgressBar(current: number, total: number, width?: number): string;
+export declare function formatOutcome(outcome: "success" | "partial" | "failed"): string;
+export declare function formatDuration(minutes: number): string;
+export declare function formatElapsed(startIso: string): string;
+//# sourceMappingURL=format.d.ts.map

package/dist/src/utils/format.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"format.d.ts","sourceRoot":"","sources":["../../../src/utils/format.ts"],"names":[],"mappings":"AAGA,wBAAgB,gBAAgB,CAAC,OAAO,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,EAAE,KAAK,SAAK,GAAG,MAAM,CAKnF;AAED,wBAAgB,aAAa,CAAC,OAAO,EAAE,SAAS,GAAG,SAAS,GAAG,QAAQ,GAAG,MAAM,CAY/E;AAED,wBAAgB,cAAc,CAAC,OAAO,EAAE,MAAM,GAAG,MAAM,CAKtD;AAED,wBAAgB,aAAa,CAAC,QAAQ,EAAE,MAAM,GAAG,MAAM,CAOtD"}

package/dist/src/utils/format.js

@@ -0,0 +1,39 @@
+const FILLED = "█";
+const EMPTY = "░";
+export function buildProgressBar(current, total, width = 10) {
+    if (total === 0)
+        return `[${EMPTY.repeat(width)}] 0/0`;
+    const filledCount = Math.min(Math.round((current / total) * width), width);
+    const emptyCount = width - filledCount;
+    return `[${FILLED.repeat(filledCount)}${EMPTY.repeat(emptyCount)}] ${current}/${total}`;
+}
+export function formatOutcome(outcome) {
+    const icons = {
+        success: "✅",
+        partial: "⚠️",
+        failed: "❌",
+    };
+    const labels = {
+        success: "Completed successfully",
+        partial: "Partially completed",
+        failed: "Failed",
+    };
+    return `${icons[outcome]} ${labels[outcome]}`;
+}
+export function formatDuration(minutes) {
+    if (minutes < 60)
+        return `~${minutes} min`;
+    const h = Math.floor(minutes / 60);
+    const m = minutes % 60;
+    return m > 0 ? `~${h}h ${m}min` : `~${h}h`;
+}
+export function formatElapsed(startIso) {
+    const ms = Date.now() - new Date(startIso).getTime();
+    const totalSeconds = Math.floor(ms / 1000);
+    if (totalSeconds < 60)
+        return `${totalSeconds}s`;
+    const minutes = Math.floor(totalSeconds / 60);
+    const seconds = totalSeconds % 60;
+    return seconds > 0 ? `${minutes}m ${seconds}s` : `${minutes}m`;
+}
+//# sourceMappingURL=format.js.map

package/dist/src/utils/format.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"format.js","sourceRoot":"","sources":["../../../src/utils/format.ts"],"names":[],"mappings":"AAAA,MAAM,MAAM,GAAG,GAAG,CAAC;AACnB,MAAM,KAAK,GAAG,GAAG,CAAC;AAElB,MAAM,UAAU,gBAAgB,CAAC,OAAe,EAAE,KAAa,EAAE,KAAK,GAAG,EAAE;IACzE,IAAI,KAAK,KAAK,CAAC;QAAE,OAAO,IAAI,KAAK,CAAC,MAAM,CAAC,KAAK,CAAC,OAAO,CAAC;IACvD,MAAM,WAAW,GAAG,IAAI,CAAC,GAAG,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC,OAAO,GAAG,KAAK,CAAC,GAAG,KAAK,CAAC,EAAE,KAAK,CAAC,CAAC;IAC3E,MAAM,UAAU,GAAG,KAAK,GAAG,WAAW,CAAC;IACvC,OAAO,IAAI,MAAM,CAAC,MAAM,CAAC,WAAW,CAAC,GAAG,KAAK,CAAC,MAAM,CAAC,UAAU,CAAC,KAAK,OAAO,IAAI,KAAK,EAAE,CAAC;AAC1F,CAAC;AAED,MAAM,UAAU,aAAa,CAAC,OAAyC;IACrE,MAAM,KAAK,GAAmC;QAC5C,OAAO,EAAE,GAAG;QACZ,OAAO,EAAE,IAAI;QACb,MAAM,EAAE,GAAG;KACZ,CAAC;IACF,MAAM,MAAM,GAAmC;QAC7C,OAAO,EAAE,wBAAwB;QACjC,OAAO,EAAE,qBAAqB;QAC9B,MAAM,EAAE,QAAQ;KACjB,CAAC;IACF,OAAO,GAAG,KAAK,CAAC,OAAO,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,EAAE,CAAC;AAChD,CAAC;AAED,MAAM,UAAU,cAAc,CAAC,OAAe;IAC5C,IAAI,OAAO,GAAG,EAAE;QAAE,OAAO,IAAI,OAAO,MAAM,CAAC;IAC3C,MAAM,CAAC,GAAG,IAAI,CAAC,KAAK,CAAC,OAAO,GAAG,EAAE,CAAC,CAAC;IACnC,MAAM,CAAC,GAAG,OAAO,GAAG,EAAE,CAAC;IACvB,OAAO,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,CAAC,GAAG,CAAC;AAC7C,CAAC;AAED,MAAM,UAAU,aAAa,CAAC,QAAgB;IAC5C,MAAM,EAAE,GAAG,IAAI,CAAC,GAAG,EAAE,GAAG,IAAI,IAAI,CAAC,QAAQ,CAAC,CAAC,OAAO,EAAE,CAAC;IACrD,MAAM,YAAY,GAAG,IAAI,CAAC,KAAK,CAAC,EAAE,GAAG,IAAI,CAAC,CAAC;IAC3C,IAAI,YAAY,GAAG,EAAE;QAAE,OAAO,GAAG,YAAY,GAAG,CAAC;IACjD,MAAM,OAAO,GAAG,IAAI,CAAC,KAAK,CAAC,YAAY,GAAG,EAAE,CAAC,CAAC;IAC9C,MAAM,OAAO,GAAG,YAAY,GAAG,EAAE,CAAC;IAClC,OAAO,OAAO,GAAG,CAAC,CAAC,CAAC,CAAC,GAAG,OAAO,KAAK,OAAO,GAAG,CAAC,CAAC,CAAC,GAAG,OAAO,GAAG,CAAC;AACjE,CAAC"}

package/openclaw.plugin.json

@@ -0,0 +1,21 @@
+{
+  "id": "claw-doing",
+  "name": "Claw Doing",
+  "description": "Real-time progress tracking for long-running OpenClaw agent tasks. See what your agent is doing — right now.",
+  "configSchema": {
+    "type": "object",
+    "properties": {
+      "storageDir": {
+        "type": "string",
+        "description": "Custom directory for task state files. Defaults to ~/.openclaw/workspace/claw-doing/tasks/"
+      },
+      "progressBarWidth": {
+        "type": "number",
+        "description": "Width of the progress bar in characters. Defaults to 10.",
+        "minimum": 5,
+        "maximum": 20
+      }
+    },
+    "additionalProperties": false
+  }
+}

package/package.json

@@ -0,0 +1,60 @@
+{
+  "name": "claw-doing",
+  "version": "0.1.0",
+  "type": "module",
+  "description": "Real-time progress tracking for long-running OpenClaw agent tasks. See what your agent is doing — right now.",
+  "keywords": [
+    "openclaw",
+    "openclaw-plugin",
+    "agent",
+    "progress",
+    "task-tracking",
+    "long-running"
+  ],
+  "license": "MIT",
+  "author": "seekcontext",
+  "repository": {
+    "type": "git",
+    "url": "https://github.com/seekcontext/claw-doing.git"
+  },
+  "homepage": "https://github.com/seekcontext/claw-doing#readme",
+  "bugs": {
+    "url": "https://github.com/seekcontext/claw-doing/issues"
+  },
+  "openclaw": {
+    "extensions": [
+      "./dist/index.js"
+    ]
+  },
+  "main": "./dist/index.js",
+  "types": "./dist/index.d.ts",
+  "exports": {
+    ".": {
+      "import": "./dist/index.js",
+      "types": "./dist/index.d.ts"
+    }
+  },
+  "files": [
+    "dist",
+    "bootstrap",
+    "SKILL.md",
+    "openclaw.plugin.json"
+  ],
+  "scripts": {
+    "build": "tsc",
+    "dev": "tsc --watch",
+    "typecheck": "tsc --noEmit",
+    "test": "vitest run",
+    "prepublishOnly": "npm run build"
+  },
+  "dependencies": {
+    "@sinclair/typebox": "^0.34.49"
+  },
+  "peerDependencies": {
+    "openclaw": ">=2026.3.0"
+  },
+  "devDependencies": {
+    "typescript": "^5.9.3",
+    "vitest": "^3.2.4"
+  }
+}