agentvibes 5.9.0 → 5.10.1

package/.claude/github-star-reminder.txt

@@ -1 +1 @@
-20260516
+20260525

package/.claude/hooks/play-tts-piper.sh

@@ -73,10 +73,8 @@ TEXT=$(printf '%s' "$TEXT" | perl -CSD -pe '
   s/^\s*[-]\s*//g;                # list dashes
 ')
 
-# Source voice manager and language manager
-# Use readlink -f to handle symlinks correctly
-SCRIPT_PATH="$(readlink -f "${BASH_SOURCE[0]}")"
-SCRIPT_DIR="$(dirname "$SCRIPT_PATH")"
+# cd-based resolution works on macOS (BSD readlink lacks -f) and Linux alike
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd -P)"
 source "$SCRIPT_DIR/piper-voice-manager.sh"
 source "$SCRIPT_DIR/language-manager.sh"
 source "$SCRIPT_DIR/audio-cache-utils.sh"
@@ -226,11 +224,19 @@ if [[ -z "$TEXT" ]]; then
 fi
 
 # Augment PATH for non-interactive shells (pipx installs to ~/.local/bin which
-# interactive shells get via .bashrc/.zshrc, but Bash tool calls skip profile)
+# interactive shells get via .bashrc/.zshrc, but Bash tool calls skip profile).
+# Mac: add both Apple Silicon (/opt/homebrew) and Intel (/usr/local) Homebrew locations.
 export PATH="$HOME/.local/bin:$HOME/.local/share/pipx/venvs/piper-tts/bin:$PATH"
+if [[ "$(uname -s 2>/dev/null)" == "Darwin" ]]; then
+  export PATH="/opt/homebrew/bin:/usr/local/bin:$PATH"
+fi
+
+# Resolve explicit piper binary path — avoids bare `piper` invocation failing when
+# PATH augmentation above hasn't propagated into nested subshells.
+PIPER_BIN=$(command -v piper 2>/dev/null || echo "")
 
 # Check if Piper is installed
-if ! command -v piper &> /dev/null; then
+if [[ -z "$PIPER_BIN" ]]; then
   echo "❌ Error: Piper TTS not installed"
   echo "Install with: pipx install piper-tts"
   echo "Or run: .claude/hooks/piper-installer.sh"
@@ -309,6 +315,8 @@ else
 fi
 
 mkdir -p "$AUDIO_DIR"
+# Normalize to canonical path (handles Git Bash /tmp→/c/Users/..., macOS /var→/private/var)
+AUDIO_DIR=$(cd "$AUDIO_DIR" && pwd -P)
 _tmp=$(mktemp "$AUDIO_DIR/tts-XXXXXX"); TEMP_FILE="${_tmp}.wav"; mv "$_tmp" "$TEMP_FILE"
 
 # @function get_speech_rate
@@ -380,6 +388,10 @@ get_speech_rate() {
 
 SPEECH_RATE=$(get_speech_rate)
 
+# Ensure piper log directory exists so stderr redirect never silently fails
+_PIPER_LOG_DIR="${AGENTVIBES_LOG_DIR:-$HOME/.local/state/agentvibes/logs}"
+mkdir -p "$_PIPER_LOG_DIR" 2>/dev/null || true
+
 # @function synthesize_with_piper
 # @intent Generate speech using Piper TTS
 # @why Provides free, offline TTS alternative
@@ -391,10 +403,10 @@ SPEECH_RATE=$(get_speech_rate)
 if [[ -n "${SPEAKER_ID:-}" ]]; then
   # Multi-speaker voice: Pass speaker ID
   # SECURITY: Use printf instead of echo for pipe safety (#134)
-  printf '%s\n' "$TEXT" | piper --model "$VOICE_PATH" --speaker "$SPEAKER_ID" --length-scale "$SPEECH_RATE" --sentence-silence 2.0 --output_file "$TEMP_FILE" 2>/dev/null
+  printf '%s\n' "$TEXT" | "$PIPER_BIN" --model "$VOICE_PATH" --speaker "$SPEAKER_ID" --length-scale "$SPEECH_RATE" --sentence-silence 2.0 --output_file "$TEMP_FILE" 2>>"$_PIPER_LOG_DIR/piper.log"
 else
   # Single-speaker voice
-  printf '%s\n' "$TEXT" | piper --model "$VOICE_PATH" --length-scale "$SPEECH_RATE" --sentence-silence 2.0 --output_file "$TEMP_FILE" 2>/dev/null
+  printf '%s\n' "$TEXT" | "$PIPER_BIN" --model "$VOICE_PATH" --length-scale "$SPEECH_RATE" --sentence-silence 2.0 --output_file "$TEMP_FILE" 2>>"$_PIPER_LOG_DIR/piper.log"
 fi
 
 if [[ ! -f "$TEMP_FILE" ]] || [[ ! -s "$TEMP_FILE" ]]; then
@@ -538,11 +550,6 @@ AUDIO_DIR="${TEMP_FILE%/*}"
 WRITE_LOCK_FILE="$AUDIO_DIR/$(basename "$TEMP_FILE" .wav).lock"
 touch "$WRITE_LOCK_FILE"
 
-# Get audio duration for proper lock timing
-DURATION=$(ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 "$TEMP_FILE" 2>/dev/null)
-DURATION=${DURATION%.*}  # Round to integer
-DURATION=${DURATION:-1}   # Default to 1 second if detection fails
-
 # Get audio duration for proper lock timing
 DURATION=$(ffprobe -v error -show_entries format=duration -of default=noprint_wrappers=1:nokey=1 "$TEMP_FILE" 2>/dev/null || true)
 DURATION=${DURATION%.*}  # Round to integer

package/.claude/hooks/requirements.txt

@@ -1,6 +1,6 @@
-# AgentVibes TTS Hooks Requirements
-# Install with: pip install -r requirements.txt
-
-# Translation support for multi-language TTS and learning mode
-deep-translator>=1.11.4
-langdetect>=1.0.9
+# AgentVibes TTS Hooks Requirements
+# Install with: pip install -r requirements.txt
+
+# Translation support for multi-language TTS and learning mode
+deep-translator>=1.11.4
+langdetect>=1.0.9

package/.claude/hooks/soprano-gradio-synth.py

@@ -1,139 +1,139 @@
-#!/usr/bin/env python3
-#
-# File: .claude/hooks/soprano-gradio-synth.py
-#
-# AgentVibes - Finally, your AI Agents can Talk Back!
-# Website: https://agentvibes.org
-# Repository: https://github.com/paulpreibisch/AgentVibes
-#
-# Co-created by Paul Preibisch with Claude AI
-# Copyright (c) 2025 Paul Preibisch
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-#
-# ---
-#
-# Soprano Gradio WebUI synthesizer helper.
-# Calls the Soprano WebUI's Gradio API and saves the result as a WAV file.
-# Uses only Python stdlib (json, sys, urllib) — no extra dependencies.
-#
-# Usage: python3 soprano-gradio-synth.py "text to speak" output.wav [port]
-#
-"""
-Soprano Gradio WebUI synthesizer helper for AgentVibes.
-
-Calls the Soprano WebUI's Gradio API via the Server-Sent Events (SSE) protocol
-and downloads the generated audio as a WAV file.
-
-Flow:
-  1. Submit generation request → get event_id
-  2. Poll SSE stream for audio file URL
-  3. Download WAV file to output path
-
-See: https://github.com/ekwek1/soprano
-"""
-import json
-import sys
-import urllib.request
-import urllib.error
-
-
-def synth(text: str, output_path: str, port: int = 7860) -> None:
-    base = f"http://127.0.0.1:{port}"
-
-    # Step 1: Submit generation request
-    # Args: text, temperature, top_p, repetition_penalty, chunk_size, streaming
-    payload = json.dumps({
-        "data": [text, 0.0, 0.95, 1.2, 1, False]
-    }).encode()
-
-    event_id = submit_request(base, payload)
-
-    # Step 2: Poll SSE stream for audio file URL
-    audio_url = poll_for_result(base, event_id)
-
-    # Step 3: Download the audio file
-    download_file(audio_url, output_path)
-
-
-def submit_request(base: str, payload: bytes) -> str:
-    """Submit generation request to Gradio API, return event_id."""
-    for api_base in ["/gradio_api/call", "/call"]:
-        url = f"{base}{api_base}/generate_speech"
-        req = urllib.request.Request(
-            url,
-            data=payload,
-            headers={"Content-Type": "application/json"},
-        )
-        try:
-            with urllib.request.urlopen(req, timeout=30) as resp:
-                return json.loads(resp.read())["event_id"]
-        except urllib.error.URLError:
-            continue
-
-    raise RuntimeError("Could not reach Soprano WebUI API")
-
-
-def poll_for_result(base: str, event_id: str) -> str:
-    """Poll SSE endpoint until audio file URL is returned."""
-    for api_base in ["/gradio_api/call", "/call"]:
-        url = f"{base}{api_base}/generate_speech/{event_id}"
-        req = urllib.request.Request(url)
-        try:
-            with urllib.request.urlopen(req, timeout=120) as resp:
-                for raw_line in resp:
-                    line = raw_line.decode("utf-8", errors="replace").strip()
-                    if not line.startswith("data: "):
-                        continue
-                    try:
-                        parsed = json.loads(line[6:])
-                    except json.JSONDecodeError:
-                        continue
-                    # Response format: [{"path": "...", "url": "...", ...}, "status string"]
-                    if isinstance(parsed, list) and len(parsed) >= 1:
-                        audio = parsed[0]
-                        if isinstance(audio, dict) and "url" in audio:
-                            return audio["url"]
-            break
-        except urllib.error.URLError:
-            continue
-
-    raise RuntimeError("No audio URL in Soprano response")
-
-
-def download_file(url: str, output_path: str) -> None:
-    """Download audio file from Gradio file server."""
-    req = urllib.request.Request(url)
-    with urllib.request.urlopen(req, timeout=30) as resp:
-        with open(output_path, "wb") as f:
-            while True:
-                chunk = resp.read(8192)
-                if not chunk:
-                    break
-                f.write(chunk)
-
-
-if __name__ == "__main__":
-    if len(sys.argv) < 3:
-        print(f"Usage: {sys.argv[0]} \"text\" output.wav [port]", file=sys.stderr)
-        sys.exit(1)
-
-    text = sys.argv[1]
-    output = sys.argv[2]
-    port = int(sys.argv[3]) if len(sys.argv) > 3 else 7860
-
-    try:
-        synth(text, output, port)
-    except Exception as e:
-        print(f"Error: {e}", file=sys.stderr)
-        sys.exit(1)
+#!/usr/bin/env python3
+#
+# File: .claude/hooks/soprano-gradio-synth.py
+#
+# AgentVibes - Finally, your AI Agents can Talk Back!
+# Website: https://agentvibes.org
+# Repository: https://github.com/paulpreibisch/AgentVibes
+#
+# Co-created by Paul Preibisch with Claude AI
+# Copyright (c) 2025 Paul Preibisch
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# ---
+#
+# Soprano Gradio WebUI synthesizer helper.
+# Calls the Soprano WebUI's Gradio API and saves the result as a WAV file.
+# Uses only Python stdlib (json, sys, urllib) — no extra dependencies.
+#
+# Usage: python3 soprano-gradio-synth.py "text to speak" output.wav [port]
+#
+"""
+Soprano Gradio WebUI synthesizer helper for AgentVibes.
+
+Calls the Soprano WebUI's Gradio API via the Server-Sent Events (SSE) protocol
+and downloads the generated audio as a WAV file.
+
+Flow:
+  1. Submit generation request → get event_id
+  2. Poll SSE stream for audio file URL
+  3. Download WAV file to output path
+
+See: https://github.com/ekwek1/soprano
+"""
+import json
+import sys
+import urllib.request
+import urllib.error
+
+
+def synth(text: str, output_path: str, port: int = 7860) -> None:
+    base = f"http://127.0.0.1:{port}"
+
+    # Step 1: Submit generation request
+    # Args: text, temperature, top_p, repetition_penalty, chunk_size, streaming
+    payload = json.dumps({
+        "data": [text, 0.0, 0.95, 1.2, 1, False]
+    }).encode()
+
+    event_id = submit_request(base, payload)
+
+    # Step 2: Poll SSE stream for audio file URL
+    audio_url = poll_for_result(base, event_id)
+
+    # Step 3: Download the audio file
+    download_file(audio_url, output_path)
+
+
+def submit_request(base: str, payload: bytes) -> str:
+    """Submit generation request to Gradio API, return event_id."""
+    for api_base in ["/gradio_api/call", "/call"]:
+        url = f"{base}{api_base}/generate_speech"
+        req = urllib.request.Request(
+            url,
+            data=payload,
+            headers={"Content-Type": "application/json"},
+        )
+        try:
+            with urllib.request.urlopen(req, timeout=30) as resp:
+                return json.loads(resp.read())["event_id"]
+        except urllib.error.URLError:
+            continue
+
+    raise RuntimeError("Could not reach Soprano WebUI API")
+
+
+def poll_for_result(base: str, event_id: str) -> str:
+    """Poll SSE endpoint until audio file URL is returned."""
+    for api_base in ["/gradio_api/call", "/call"]:
+        url = f"{base}{api_base}/generate_speech/{event_id}"
+        req = urllib.request.Request(url)
+        try:
+            with urllib.request.urlopen(req, timeout=120) as resp:
+                for raw_line in resp:
+                    line = raw_line.decode("utf-8", errors="replace").strip()
+                    if not line.startswith("data: "):
+                        continue
+                    try:
+                        parsed = json.loads(line[6:])
+                    except json.JSONDecodeError:
+                        continue
+                    # Response format: [{"path": "...", "url": "...", ...}, "status string"]
+                    if isinstance(parsed, list) and len(parsed) >= 1:
+                        audio = parsed[0]
+                        if isinstance(audio, dict) and "url" in audio:
+                            return audio["url"]
+            break
+        except urllib.error.URLError:
+            continue
+
+    raise RuntimeError("No audio URL in Soprano response")
+
+
+def download_file(url: str, output_path: str) -> None:
+    """Download audio file from Gradio file server."""
+    req = urllib.request.Request(url)
+    with urllib.request.urlopen(req, timeout=30) as resp:
+        with open(output_path, "wb") as f:
+            while True:
+                chunk = resp.read(8192)
+                if not chunk:
+                    break
+                f.write(chunk)
+
+
+if __name__ == "__main__":
+    if len(sys.argv) < 3:
+        print(f"Usage: {sys.argv[0]} \"text\" output.wav [port]", file=sys.stderr)
+        sys.exit(1)
+
+    text = sys.argv[1]
+    output = sys.argv[2]
+    port = int(sys.argv[3]) if len(sys.argv) > 3 else 7860
+
+    try:
+        synth(text, output, port)
+    except Exception as e:
+        print(f"Error: {e}", file=sys.stderr)
+        sys.exit(1)