Conversation
added 30 commits
April 3, 2026 20:43
- Issue parcadei#169: Replace COUNT(*) FILTER(WHERE) with SUM(CASE WHEN...) in BTQL query - Issue parcadei#168: Use spawn context for ProcessPoolExecutor on macOS - Issue parcadei#152: Fix relative path resolution using input.cwd in tldr-read-enforcer - Issue parcadei#156/157: Port ensureMemoryDaemon() from Python to TypeScript session-start hook
/parcadei#155) - Add _copy_dotfile_scripts() to copy root-level .py/.sh files from .claude/scripts/ - Called from both install_opc_integration() and install_opc_integration_symlink() - Add missing 'Installed N scripts' print line in wizard for both install paths - Files now deployed: status.py, recall_learnings.py, tldr_stats.py
- Add fish shell detection (use 'set -gx' syntax, ~/.config/fish/config.fish) - Create ~/.config/fish/ directory if it doesn't exist - Remove CLAUDE_OPC_DIR from shell config on uninstall (issue parcadei#147)
- Remove 4 archived skills from active list (async-repl-protocol, router-first-architecture, search-hierarchy, search-tools) - Rename 7 underscore directories to kebab-case (continuity_ledger, create_handoff, describe_pr, implement_task, implement_plan, resume_handoff, system_overview) - Fix name: field in implement-task and implement-plan SKILL.md - Remove outdated model pin claude-opus-4-5-20251101 from discovery-interview and research skills
…dei#144) - Add Step 11/15: qlty code quality tool (70+ linters, 40+ languages) - Add Step 12/15: ast-grep AST-based code search and refactoring - Update TLDR to Step 13/15, Diagnostics to Step 14/15, Loogle to Step 15/15 - Total steps increased from 13 to 15
…parcadei#146) - Add Firecrawl API key prompt (web scraping) → FIRECRAWL_API_KEY env var - Add Morph API key prompt (data extraction) → MORPH_API_KEY env var - Unlocks: firecrawl-scrape, research-agent, research-external skills
- New hook blocks scripts/(mcp|core)/ commands outside OPC directory - Prevents ModuleNotFoundError when uv run misses pyproject.toml - Returns corrected command with cd prefix suggestion - Fixes parcadei#148
- Add CLAUDE_CC_DIR alongside CLAUDE_OPC_DIR in wizard.py setup - Update 8 agents to use CLAUDE_CC_DIR for project paths - Update 8 skills to use CLAUDE_CC_DIR for project paths - References paradei PR parcadei#138
- Removes timeout from npm install, npm build - Removes timeout from wait_for_services() - Removes timeout from uv sync, uv tool install, cargo install - Removes timeout from lake build - Removes timeout from model download subprocess calls - Removes related comments
- uv tool install qlty: 120s -> 300s (5 min for first-time install) - curl install script: 30s -> 60s - curl | sh qlty install: 60s -> 120s - cargo install ast-grep: 300s -> 600s (10 min for build from source) - git clone loogle: 120s -> 600s (2GB repo download) These prevent premature timeout failures while still protecting against infinite hangs.
When user selects 'embedded' or 'sqlite' database mode, skip: - Step 6: Docker stack (no need for Docker containers) - Step 7: Database migrations (handled automatically by embedded or not needed for sqlite)
After setting up embedded PostgreSQL: 1. Start the pgserver 2. Run docker/init-schema.sql via run_migrations_direct() 3. Show warnings for optional extensions that failed
start_embedded_postgres() now accepts venv_path parameter and uses venv's Python executable to run pgserver as subprocess when pgserver is not available in the current Python environment.
Instead of running pgserver as subprocess, now adds venv's site-packages to sys.path before importing pgserver module.
Since global Python (3.14) cannot import modules compiled for venv Python (3.12), use subprocess to run pgserver startup script with the correct venv Python interpreter.
The docker/ directory is at repo root, not opc/ root.
Embedded PostgreSQL setup is now simpler - just creates venv. User handles server start and schema migration manually.
qlty is a standalone binary installed to ~/.qlty/bin/qlty, not a uv tool.
- ast-grep: check with shutil.which before install - tldr: check if llm-tldr is installed (not tldr-pages)
- qlty: just check shutil.which('qlty')
- tldr: check shutil.which and verify it's llm-tldr
- ast-grep: just check shutil.which('ast-grep')
Model download via subprocess doesn't work because tldr module is installed in uv tool environment, not system Python. User should run 'tldr semantic index .' manually after setup.
Instead of sys.executable + module import, use 'uv run tldr semantic download-model' which works regardless of where tldr is installed.
There is no 'download-model' command - model downloads automatically when running 'tldr semantic index'. Using correct command now.
added 3 commits
April 4, 2026 00:07
Using sys.executable fails because tldr module is in uv tool environment, not system Python. Using 'uv run python' instead.
Add detailed docstring explaining: - Output format with examples - Color coding scheme - All components (tokens, context%, branch, S/U/A counts, goal, now) - Temporary file locations
CONTEXT: {tokens} {context}% | GIT: {branch} S:{staged} U:{unstaged} A:{added} | GOAL: {goal} → NOW: {now}
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Fixes for multiple issues from the upstream paradei/Continuous-Claude-v3 repository:
Bug Fixes
COUNT(*) FILTER(WHERE...)withSUM(CASE WHEN...)in BTQL queryspawncontext forProcessPoolExecutoron macOStldr-read-enforcerhookensureMemoryDaemon()from Python to TypeScript.claude/scripts/root files not being copied during installCLAUDE_OPC_DIRon uninstallSkill Reference Fixes (parcadei#158)
name:field in implement-task and implement-planclaude-opus-4-5-20251101Wizard Enhancements
Commits
3b968e0- fix: resolve multiple issues (BTQL, spawn, path, memory daemon)f506739- fix: copy .claude/scripts/ root files during install77d3316- fix: add fish shell support and CLAUDE_OPC_DIR cleanup0284129- fix: resolve skill reference issues22db469- feat: add qlty and ast-grep to wizarde2b3c61- feat: add Firecrawl and Morph API keysTesting