Skip to main content

bisos.startAiActivity: Install AI Collaborative Development Templates

Project description

Overview

bisos.startAiActivity installs the ByStar AI collaborative development template set (CLAUDE.md, AI-AGENTS.org, AI-WORKFLOW.org, AI-Activity.org, AI-DevStatus.org, AI-WorkPlan.org) into any project directory or git repo root, wiring it for collaborative development with Claude Code.

bisos.startAiActivity is a python package that uses the PyCS-Framework. It provides the startAiActivity.cs command for installing and managing AI collaborative development templates.

Table of Contents TOC

Part of BISOS — ByStar Internet Services Operating System

Layered on top of Debian, BISOS (By* Internet Services Operating System) is a unified and universal framework for developing both internet services and software-service continuums that use internet services. See Bootstrapping ByStar, BISOS and Blee for information about getting started with BISOS.

BISOS is a foundation for The Libre-Halaal ByStar Digital Ecosystem which is described as a cure for losses of autonomy and privacy in a book titled: Nature of Polyexistentials

bisos.startAiActivity is part of BISOS. It is a standalone package that can be used independently of the full BISOS environment.

bisos.startAiActivity is a Command-Only PyCS Facility

bisos.startAiActivity is a command-line tool. It is a PyCS single-unit command service. PyCS is a framework that converges development of CLI tools and services. PyCS is an alternative to FastAPI, Typer and Click.

bisos.startAiActivity uses the PyCS-Framework to:

  1. Install AI collaborative development templates (symlinks and safe-copies) into any project directory or git repo root.

  2. Expand b:ai:file/particulars org-mode dynamic blocks in installed files using pure Python, without requiring Emacs or bx-dblock.

  3. Manage persistent user-config parameters (e.g. templates base path) via ~/.config/bisos/startAiActivity.cs/.

The core of PyCS-Framework is the bisos.b package (the PyCS-Foundation). See bisos.b for an overview.

Post-Installation Setup

After installing, configure the templates base directory. If /bisos/apps/defaults/ai-templates exists on your system it is used as the default automatically. Otherwise, set it explicitly:

startAiActivity.cs -i userConfig_set --parName=templates --parValue=/path/to/ai-templates

Verify the setting:

startAiActivity.cs -i userConfig_get --parName=templates

The AI templates are maintained at: https://github.com/bisos/defaults/tree/master/ai-templates

Installation

The sources for the bisos.startAiActivity pip package are maintained at: https://github.com/bisos-pip/startAiActivity.

The bisos.startAiActivity pip package is available at PYPI as https://pypi.org/project/bisos.startAiActivity

You can install bisos.startAiActivity with pip or pipx.

Installation With pip

If you need access to bisos.startAiActivity as a python module, you can install it with pip:

pip install bisos.startAiActivity

Installation With pipx

If you only need access to bisos.startAiActivity on command-line, you can install it with pipx:

pipx install bisos.startAiActivity

The following commands are made available:

  • startAiActivity.cs — installs AI collaborative development templates

Post-Installation Basic Testing

After installation, run:

startAiActivity.cs
startAiActivity.cs -i userConfig_get --parName=templates

Usage

Set Templates Base Directory

startAiActivity.cs -i userConfig_set --parName=templates --parValue=/path/to/ai-templates

Install Templates Into a Project

Install into the current directory:

startAiActivity.cs -i initiate --activity=bisos-pip

Install at the git repo root:

startAiActivity.cs -i initiate --root=repo --activity=bisos-pip

Available activities match subdirectories under the templates base (e.g. bisos-pip, bisos-core, xu-single, blee-panels, lcnt). Run startAiActivity.cs -i examples to see all available activities.

initiate installs the following into the target directory: symlinks for the constant files (CLAUDE.md, AI-AGENTS.org, AI-WORKFLOW.org, AI-Activity.org); safe-copies of the editable files (AI-DevStatus.org, AI-WorkPlan.org, with b:ai:file/particulars dblocks expanded); and symlinks under .claude/ for settings.json, commands/, and skills/ (when the activity ships a _skills/ directory). See the Skills subsection below.

Install a Subproject Overlay

For repos with nested loosely-related subprojects that each want their own work plan and dev status without duplicating the invariant files (AI-AGENTS.org, AI-WORKFLOW.org, .claude/), use initiateSub:

startAiActivity.cs -i initiateSub --activity=bisos-pip

This installs only the subproject-specific state — a slim CLAUDE.md (symlinked to mother/initiateSub/CLAUDE.md), AI-Activity.org symlink, and safe-copies of AI-DevStatus.org and AI-WorkPlan.org. It does NOT install AI-AGENTS.org, AI-WORKFLOW.org, or .claude/ — those are inherited from an ancestor directory via Claude Code’s CLAUDE.md walk-up.

Preconditions (initiateSub refuses if either fails):

  1. An ancestor directory must have been initiate-d already (initiateSub walks up looking for a startAiActivity-signature CLAUDE.md symlink).

  2. The target directory must not already have a CLAUDE.md.

The --activity flag is required — a subproject may be a different activity from its parent (e.g., a blee-panels subproject inside a bisos-pip repo).

Nested initiateSub is supported: a subproject’s CLAUDE.md satisfies the walk-up precondition for a sub-subproject.

aiSuspend / aiResume / deClaudify all work correctly on subproject overlays. aiResume detects sub vs. base mode from the stashed CLAUDE.md.dormant symlink target.

Skills

When an activity ships a _skills/ directory at its templates root, initiate symlinks it to .claude/skills/ alongside the other .claude/ entries. Claude Code discovers those skills automatically and loads them on demand when the current task calls for them.

Skills are self-contained, filesystem-based packages of task-specific instructions. Each skill is a folder containing a SKILL.md file whose front-matter description tells Claude when to consult the full file. This lets a repo carry deep procedural knowledge (how to write a CS command, how to author a Blee panel, how to release to PyPI) without loading that content on every session — the short descriptions sit in context by default, and Claude reads the full skill only when it matches the task at hand.

The bisos-pip activity currently ships five skills:

  • writing-cs-commands — authoring cs.Cmnd subclasses (params, args, classHead dblock, common patterns).

  • dblock-authoring — working with poly-dblock begin: blocks, the csuListProc pattern, and dblock-based inline unit tests.

  • readme-authorship — writing README.org (repo root — the only authored README; py3/README.rst is generated by pypiProc.sh -i fullPrep and must never be hand-edited).

  • blee-panel-authorship — the standard fullUsagePanel-en.org skeleton and dblock content policy.

  • bisos-pip-release — end-to-end PyPI release checklist (linters, packaging, Test-PyPI trial, upload).

To author a new skill, add a folder under the appropriate templates directory:

  • mother/_skills/<name>/SKILL.md — available to every activity

  • <activity>/_skills/<name>/SKILL.md — activity-specific

Each SKILL.md starts with YAML front-matter (----delimited) containing name: and description: fields. The description drives when Claude loads the skill; write it as a concrete trigger (“Use when writing or modifying a PyCS cs.Cmnd subclass”) rather than a topic label.

Suspend and Resume AI Collaboration

When AI collaboration on a project needs to be paused temporarily, use aiSuspend to enter a dormant state and aiResume to restore it.

aiSuspend removes the constant symlinks (CLAUDE.md, AI-AGENTS.org, AI-WORKFLOW.org, AI-Activity.org) and the .claude/ entries, and renames the editable files AI-DevStatus.org and AI-WorkPlan.org to .dormant suffixes so that the collaboration state survives.

startAiActivity.cs -i aiSuspend

aiResume restores the .dormant files, re-installs the symlinks, and re-installs .claude/. The activity is inferred from the AI-Activity.org symlink target or from the Activity: header inside AI-WorkPlan.org.

startAiActivity.cs -i aiResume

Both commands operate on the current directory.

Remove AI Collaboration Files

To permanently remove all AI collaboration files from a project, use deClaudify. Unlike aiSuspend, this is irreversible — it deletes the editable AI-DevStatus.org and AI-WorkPlan.org files along with the symlinks and .claude/ entries.

startAiActivity.cs -i deClaudify

The command is safe: symlinks are only removed if they are actually symlinks, and copied files are only removed if they are regular files. Anything a user has created or converted manually is left intact.

Key Files

An overview of the relevant files of the bisos.startAiActivity package:

py3/bin/startAiActivity.cs

The main CS-MU (Command-Services Multi-Unit). Provides the initiate, aiSuspend, aiResume, deClaudify, userConfig_get, and userConfig_set commands. Uses bisos.b.userConfig_csu for persistent parameter management.

py3/bisos/startAiActivity/updateDblock.py

General dblock dispatch engine. Handlers register under a signature string via registerHandler(sig, fn). expandAll(filePath) does a single-pass regex substitution over the file, dispatching each begin: <signature> block to its registered handler.

py3/bisos/startAiActivity/dblock_particulars.py

Self-registering handler for the b:ai:file/particulars dblock signature. Derives the activity from the AI-Activity.org symlink and builds the companion-docs list by scanning the directory.

PyPi and Github Packaging

All bisos-pip repos in the https://github.com/bisos-pip github organization follow the same structure. They all have setup.py files driven by pypiProc.sh.

Testing

An end-to-end verification harness lives at py3/test/verify.sh. It exercises every command in the startAiActivity.cs surface — user-config read/write, initiate, deClaudify, aiSuspend / aiResume (with base-mode detection), and initiateSub (both the positive case within the repo and a negative case in an isolated /tmp/ dir that expects a EH_Problem_UsageError refusal).

Run it as:

cd py3/test
./verify.sh

The shebang is #!/bin/bash -i — this makes the shell interactive, so ~/.bashrc runs its full body and defines the lpDo alias (from ByStar’s _opDo convention) used throughout the script for uniform logging.

The harness starts from a clean state and leaves a clean state — the test directory contains only verify.sh itself before and after a successful run. Exit code 0 indicates all steps behaved as expected (including the intended refusal in the negative case).

Documentation and Blee-Panels

bisos.startAiActivity is part of ByStar Digital Ecosystem http://www.by-star.net.

This module’s primary documentation is in the form of Blee-Panels. Additional information is also available in: http://www.by-star.net/PLPC/180047

bisos.startAiActivity Blee-Panels

bisos.startAiActivity Blee-Panels are in the ./panels directory. From within Blee and BISOS these panels are accessible under the Blee “Panels” menu.

See file:./panels/bisos.startAiActivity/_nodeBase_/fullUsagePanel-en.org for a starting point.

Support

For support, criticism, comments and questions; please contact the author/maintainer

Planned Improvements

  • Upload to PyPI and test as pipx install

  • Write fullUsagePanel-en.org Blee-Panel

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

bisos_startaiactivity-0.11.tar.gz (36.0 kB view details)

Uploaded Source

File details

Details for the file bisos_startaiactivity-0.11.tar.gz.

File metadata

  • Download URL: bisos_startaiactivity-0.11.tar.gz
  • Upload date:
  • Size: 36.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.11.2

File hashes

Hashes for bisos_startaiactivity-0.11.tar.gz
Algorithm Hash digest
SHA256 1a205241bd39c061d5b1d034b585bbf0335f9ee342ed95271bb03d65c5bf74d5
MD5 1d98ab38c7a02426b194cb92d03acbff
BLAKE2b-256 064c466791b8cc9eddcf932ed07c0f29dbef548f489ff314e39a1055150fa7c9

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page