readme cleanup

AGENTS.md

@@ -54,7 +54,7 @@ Task IDs are zero-padded strings (`"01"`, `"02"`, etc.). The parser prepends `0`
 
 ## Command routing
 
-`/ralpi` with no args → plan. First token looks like a path (`@path`, `./path`, `.md`, etc.) → run. Otherwise dispatches to subcommand (`run`, `plan`, `status`, `resume`, `next`, `reset`).
+`/ralpi` with no args → plan. First token looks like a path (`@path`, `./path`, `.md`, etc.) → run. Otherwise dispatches to subcommand (`run`, `plan`, `resume`, `reset`).
 
 ## Config
 

README.md

@@ -4,75 +4,33 @@ Execute tasks from task files using DAG-based dependency resolution with persist
 
 ## Features
 
-- **DAG-based execution**: Tasks are ordered by dependencies using Kahn's algorithm
 - **Parallel batching**: Independent tasks in each batch can run concurrently
 - **Persistent progress**: Execution state saved to `.ralpi/progress.json`
 - **Reflection system**: Each task produces a reflection for downstream tasks
 - **Retry with backoff**: Failed tasks retry with exponential backoff
-- **Multiple formats**: Supports Fio README, simple checkboxes, and YAML
+- **Multiple formats**: Supports simple checkboxes, and YAML
-- **Chat progress**: Real-time progress messages in Pi chat via `pi.sendMessage`
 - **Tool usage tracking**: Detects and reports tool usage (read, write, edit, bash) from task execution
-- **Git commit capture**: Captures git commit messages and generates summaries per task
 - **Configurable timeouts**: Task-level timeouts via meta blocks, with global fallback
 - **Session saving**: Saves full task output for expandable session review
 - **Resume auto-discovery**: Automatically finds and resumes interrupted execution
-- **Custom message renderer**: Compact UI labels with expandable details in Pi TUI
 
 ## Usage
 
 ```
-/ralpi plan [task-file]   # Show execution plan
+/ralpi [task-file]        # Execute all tasks
-/ralpi run [task-file]    # Execute all tasks
+/ralpi plan               # Alias to /task-manager to plan new tasks
-/ralpi status [task-file] # Show current progress
+/ralpi resume             # Resume paused execution
-/ralpi resume [task-file] # Resume paused execution
+/ralpi reset [task-file]  # Reset progress and .ralpi directory - does not modify PRD
-/ralpi next [task-file]   # Execute next batch only
-/ralpi reset [task-file]  # Reset all progress
 ```
 
 ## Task File Formats
 
-### Fio README Format
+### Highly recommended to use the task-manager prompt for prd construction, it's output pairs perfectly
 
-```markdown
 # Project Title
 
 ## Tasks
 
-- [ ] 01 — Setup project structure -> `tasks/01-setup.md`
-- [ ] 02 — Implement auth -> `tasks/02-auth.md`
-- [ ] 03 — Build API -> `tasks/03-api.md`
-
-## Dependencies
-
-1 -> 2,3
-2 -> 3
-```
-
-#### Supported Dependency Formats
-
-The parser supports two dependency declaration styles in the `## Dependencies` section:
-
-**Arrow Notation** (recommended):
-```
-1 -> 2,3,4
-5 -> 6
-```
-This means: "Task 1 must complete before tasks 2, 3, and 4 can start."
-
-**Natural Language**:
-```
-13 depends on 17, 18, 19, 20
-14 depends on 13, 15, 16
-```
-This means: "Task 13 depends on tasks 17, 18, 19, and 20."
-
-**Parallel Groups** (informational only):
-```
-1, 2, 3, 4 can be done in parallel
-5, 6, 7, 8 can be done in parallel
-```
-Note: These lines are ignored by the parser. Use explicit dependencies to control execution order.
-
 ### Simple Checkbox Format
 
 ```markdown
@@ -96,9 +54,45 @@ tasks:
     depends_on: ["01"]
 ```
 
+## Dependencies
+
+### Arrow Notation (recommended):
+
+1 -> 2,3,4
+5 -> 6
+This means: "Task 1 must complete before tasks 2, 3, and 4 can start."
+
+### Natural Language:
+
+13 depends on 17, 18, 19, 20
+14 depends on 13, 15, 16
+
+This means: "Task 13 depends on tasks 17, 18, 19, and 20."
+
+### Parallel Groups (informational only):
+
+1, 2, 3, 4 can be done in parallel
+5, 6, 7, 8 can be done in parallel
+
+Note: These lines are ignored by the parser. Use explicit dependencies to control execution order.
+
+
+
 ## Configuration
 
-Create config files. Both are optional:
+### Task-Level Timeout
+
+You can set a timeout for individual tasks using a meta block in the task file:
+
+```markdown
+- [ ] 01: Setup project structure
+  timeout: 10m
+```
+
+Supported formats: `10m` (minutes), `600s` (seconds), `3600000` (milliseconds)
+
+
+### Config files
 
 | Scope | Path |
 |-------|------|
@@ -115,9 +109,6 @@ prompts:
   projectContext: "Additional context for all tasks"
 ```
 
-> ralpi deliberately does **not** set timeouts or retries — those are inherited
-> from Pi's own settings. Tasks run until they complete or Pi's own flow stops them.
->
 > `execution.models` uses slot-aware round-robin: with 3 models and 2 concurrent
 > tasks, only the first two models are used. The third model is only touched when
 > a third concurrent task starts. Freed model slots are reused before new ones
@@ -127,28 +118,6 @@ prompts:
 > the task automatically cycles to the next model in the list without counting it
 > as a task failure. Each model is tried once before the task is marked as failed.
 
-The keys mirror the nested structure of `RalpiConfig` in `src/types.ts`.
-
-### Precedence (highest wins)
-
-| Priority | Source |
-|----------|--------|
-| **1st** | In-memory overrides (`model`, `thinkingLevel` from parent Pi session) |
-| **2nd** | `./.ralpi/config.yaml` — project-level |
-| **3rd** | `~/.pi/ralpi/config.yaml` — global, shared across projects |
-| **4th** | `DEFAULT_CONFIG` in `src/types.ts` |
-
-### Task-Level Timeout
-
-You can set a timeout for individual tasks using a meta block in the task file:
-
-```markdown
-- [ ] 01: Setup project structure
-  timeout: 10m
-```
-
-Supported formats: `10m` (minutes), `600s` (seconds), `3600000` (milliseconds)
-
 ## State Files
 
 - `.ralpi/progress.json` - Execution progress

package.json

@@ -8,7 +8,7 @@
     "task-runner",
     "dag",
     "task-manager",
-    "ralpi-loop",
+    "ralph-loop",
     "prd"
   ],
   "author": "Michael Freno",
@@ -54,14 +54,6 @@
     "@earendil-works/pi-coding-agent": "*",
     "@earendil-works/pi-tui": "*"
   },
-  "peerDependenciesMeta": {
-    "@earendil-works/pi-coding-agent": {
-      "optional": true
-    },
-    "@earendil-works/pi-tui": {
-      "optional": true
-    }
-  },
   "publishConfig": {
     "access": "public"
   },

src/executor.ts

@@ -296,8 +296,8 @@ export async function runTask(
 	}
 
 	if (!output.success) {
-		sendChatMessage?.(`✗ ${taskHeader} — ${output.error}`);
+		// Failure reporting is handled by the caller (executeTask) to avoid
-		ctx.ui.notify(`Task ${task.id} failed: ${output.error}`, "error");
+		// duplicate messages when model failover or retry cycling is active.
 		return {
 			success: false,
 			error: output.error,
@@ -433,6 +433,7 @@ export async function executeBatch(
 			roundRobin?.release(task.id);
 			const errorMsg = error instanceof Error ? error.message : String(error);
 			progress.markFailed(task.id, errorMsg);
+			sendChatMessage?.(`✗ ${task.id} · ${task.title} — ${errorMsg}`);
 			ctx.ui.notify(`Task ${task.id} failed: ${errorMsg}`, "error");
 			break;
 		}
@@ -555,6 +556,7 @@ async function executeBatchParallel(
 				roundRobin?.release(task.id);
 				const errorMsg = error instanceof Error ? error.message : String(error);
 				progress.markFailed(task.id, errorMsg);
+				sendChatMessage?.(`✗ ${task.id} · ${task.title} — ${errorMsg}`);
 				ctx.ui.notify(`Task ${task.id} failed: ${errorMsg}`, "error");
 			}),
 		});
@@ -658,9 +660,8 @@ async function executeTask(
 					// release() would put the slot in freeSlots, then assign()
 					// would pick it right back up, getting stuck on the same model.
 					modelAttempt++;
-					ctx.ui.notify(
+					sendChatMessage?.(
-						`Task ${task.id}: model failed, trying next (${modelAttempt + 1}/${maxModelAttempts}): ${result.error}`,
+						`~ ${task.id} · ${task.title} — trying model ${modelAttempt + 1}/${maxModelAttempts} (previous: ${result.error})`,
-						"warning",
 					);
 					break; // exit retry loop, cycle to next model
 				}
@@ -668,9 +669,8 @@ async function executeTask(
 				// No more models — use normal retry logic
 				if (retries < maxRetries) {
 					retries = progress.incrementRetry(task.id);
-					ctx.ui.notify(
+					sendChatMessage?.(
-						`Retrying task ${task.id} (${retries}/${maxRetries}): ${result.error}`,
+						`~ ${task.id} · ${task.title} — retrying (${retries}/${maxRetries}): ${result.error}`,
-						"warning",
 					);
 
 					// Exponential backoff
@@ -679,6 +679,7 @@ async function executeTask(
 				} else {
 					// Max retries exceeded
 					progress.markFailed(task.id, result.error || "Unknown error");
+					sendChatMessage?.(`✗ ${task.id} · ${task.title} — ${result.error}`);
 					ctx.ui.notify(
 						`Task ${task.id} failed after ${maxRetries} retries: ${
 							result.error || "Unknown error"
@@ -691,6 +692,7 @@ async function executeTask(
 				roundRobin?.release(task.id);
 				const errorMsg = error instanceof Error ? error.message : String(error);
 				progress.markFailed(task.id, errorMsg);
+				sendChatMessage?.(`✗ ${task.id} · ${task.title} — ${errorMsg}`);
 				ctx.ui.notify(`Task ${task.id} failed: ${errorMsg}`, "error");
 				return;
 			}
@@ -703,6 +705,9 @@ async function executeTask(
 	// All models exhausted — release the slot
 	roundRobin?.release(task.id);
 	progress.markFailed(task.id, "All configured models exhausted");
+	sendChatMessage?.(
+		`✗ ${task.id} · ${task.title} — all ${maxModelAttempts} models exhausted`,
+	);
 	ctx.ui.notify(
 		`Task ${task.id} failed: all configured models exhausted`,
 		"error",