Menu
Automated software development pipeline. Scans projects, prioritizes tasks, analyzes code, and orchestrates development loops. Zero dependencies (Python stdlib only).
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Plant Unterricht, Lernangebote und individuelle Förderung ohne Berichtsgenerator oder persönliche Vorlagen.
Umbrella skill for cloud-bridged communication protocols between agents on different machines (Ping-Pong, agent-beam, listeners and future protocols). Use when coordinating work across machines via a shared sync folder or message yard.
Self-tuning cadence control loop for recurring agent scans. Use when a scheduled scan should sharpen its interval on activity and cool down on silence, without operator intervention.
Führt einen Hackathon end-to-end als Operator: von der Ausschreibung über Analyse, Ideenfindung, Bau, Beweisführung, Medienproduktion bis zur Einreichung — mit Zwischenstandsspeicher (STATE.md), aus dem jede Phase wieder aufgenommen werden kann. Der Nutzer ist human-in-the-loop: Richtungsvorgaben, Abnahmen und alle unumkehrbaren Aktionen (Public, Upload, Submit) bleiben beim Menschen.
Compose video-synced background scores from a storyline JSON using local waveform synthesis (numpy + ffmpeg). Styles: chiptune, ambient, electronic. Deterministic, offline, no cloud service.
Ordnet Büroaufgaben, Korrespondenz und Fristen ohne Bindung an eine bestimmte Office-App.
| name | dev-soft-agent |
| version | 0.1.0 |
| type | agent |
| author | BACH Team |
| created | "2026-02-21T00:00:00.000Z" |
| updated | "2026-03-12T00:00:00.000Z" |
| description | Automated software development pipeline. Scans projects, prioritizes tasks, analyzes code, and orchestrates development loops. Zero dependencies (Python stdlib only). |
| standalone | true |
| anthropic_compatible | true |
| bach_compatible | true |
| bach_origin | true |
| category | dev |
| tags | ["development","code-analysis","task-management","automation","pipeline"] |
| language | de |
| status | active |
| dependencies | {"tools":[],"services":[],"protocols":[],"python":[]} |
| provenance | {"origin":"bach","origin_path":"MODULAR_AGENTS/devSoftAgent","origin_version":"0.1.0","origin_repo":"github.com/ellmos-ai/bach","last_sync_from_origin":"2026-03-12","last_sync_to_origin":"None","local_changes_since_sync":false} |
Deutsch — Offizielle Deutsch-Version / Documento Oficial en Deutsch.
Automated software development pipeline. Extracted from BACH's ATI agent, runs fully standalone with pure Python standard library.
scripts/
config.py Configuration (scan folders, naming prefixes, weights)
project_manager.py Project scan + classification by naming convention
task_engine.py TASKS.txt parser + code scanner (TODO/FIXME)
code_analyzer.py Static analysis (LOC, imports, classes, functions)
dev_loop.py Orchestrator (DevLoop)
policies/
naming.py snake_case / PascalCase / SCREAMING_SNAKE validation
encoding.py UTF-8 enforcement + BOM detection
paths.py Hardcoded path detection
prompt_templates/
task_prompt.txt LLM prompt for task processing
review_prompt.txt LLM prompt for code review
analysis_prompt.txt LLM prompt for project analysis
from scripts.dev_loop import DevLoop
from scripts.config import Config
config = Config()
loop = DevLoop(config)
# Scan projects (Deutsch)
projects = loop.scan_projects()
# Select project (weighted random selection by naming convention) (Deutsch)
project = loop.select_project()
# Analyze code (Deutsch)
analysis = loop.analyze_project()
print(f"{analysis.total_loc} LOC, {analysis.todo_count} TODOs")
# Load and prioritize tasks (Deutsch)
tasks = loop.get_tasks()
for task in tasks:
print(f"[{task.task_type.name}] {task.description} (Prio: {task.priority})")
# Complete dev session (Deutsch)
result = loop.run_session()
loop.save_session()
cd scripts
python -m devSoftAgent scan ~/projects
python -m devSoftAgent select
python -m devSoftAgent analyze /path/to/project
python -m devSoftAgent tasks /path/to/project
python -m devSoftAgent session --project my-project
python -m devSoftAgent status
Projects are classified based on their folder name:
| Prefix | Label | Weight | Meaning |
|---|---|---|---|
RDY | Ready | 1.0 | Highest priority |
RDY_FAST | Fast Ready | 0.5 | Quick to complete |
FAST | Fast | 0.33 | Small task |
DEV | Development | 0.17 | In development |
REL | Released | 0.0 | Done, no work needed |
ARC | Archived | 0.0 | Archived |
Weight determines the probability in random selection.
# TASKS - ProjectName (Deutsch)
# As of: 2026-03-12 (Deutsch)
## OPEN
- [ ] [BUG] Description of the bug
- [ ] [FEATURE] New feature
## IN PROGRESS
- [-] [REFACTOR] Code restructuring
## DONE
- [x] [BUG] Fixed bug -- DONE 2026-03-01
Quality policies that can be automatically checked against code: