testing-workflows

name	testing-workflows
description	Test bot functionality locally using CLI runner, message simulation, and workflow validation. Trigger on "test", "CLI runner", "message simulation", "bot testing", "workflow validation", "pytest".
version	2.0.1
allowed-tools	Bash, Read, Write

Testing Workflows Skill

Test bot functionality locally using CLI tools, message simulation, and unit tests.

CLI Summary Runner

The CLI runner tests the full URL processing pipeline without Telegram credentials. Telegram credentials (API_ID, API_HASH, BOT_TOKEN) are NOT required -- the CLI generates stub credentials automatically.

Basic Usage

# Summarize a single URL
python -m app.cli.summary --url https://example.com/article

# With custom output and debug logging
python -m app.cli.summary \
  --url https://example.com/article \
  --json-path summary.json \
  --log-level DEBUG

# Auto-accept multiple URLs
python -m app.cli.summary \
  --url "https://example.com/1 https://example.com/2" \
  --accept-multiple

# Simulate message text (like Telegram input)
python -m app.cli.summary "/summarize https://example.com/article"

CLI Features

• Full pipeline: URL normalization -> scraper chain -> LLM -> JSON validation
• Deduplication: Respects dedupe_hash (won't re-crawl same URL)
• Insights generation: Optional advanced analysis with retry logic
• JSON repair: Handles malformed LLM output
• Correlation IDs: Generates unique IDs for tracing

Testing Specific Workflows

Standalone test scripts are bundled with this skill in ./scripts/. Run them directly:

.venv/bin/python .claude/skills/testing-workflows/scripts/test-url-normalization.py
.venv/bin/python .claude/skills/testing-workflows/scripts/test-summary-validation.py
.venv/bin/python .claude/skills/testing-workflows/scripts/test-language-detection.py
.venv/bin/python .claude/skills/testing-workflows/scripts/test-access-control.py

Simulating Telegram Messages

Message Models

Telegram adapters receive Telethon message objects. Follow existing fixtures and MagicMock/SimpleNamespace shapes in tests/adapters/telegram/ and tests/test_command_dispatcher.py; do not invent a parallel Telegram message model.

Test Message Router

For routing logic tests, instantiate MessageRouter with mocked dependencies and call it directly. See app/adapters/telegram/message_router.py for the constructor signature and existing router tests under tests/adapters/telegram/ for setup patterns.

End-to-End URL Flow

# Run full pipeline
python -m app.cli.summary \
  --url https://example.com/article \
  --json-path test_output.json \
  --log-level DEBUG

# Verify output
python -m json.tool test_output.json

# Verify in database
docker exec -i ratatoskr-postgres psql -U ratatoskr_app -d ratatoskr -c \
  "SELECT id, type, status, input_url
     FROM requests
    ORDER BY created_at DESC
    LIMIT 1;"

Error Handling

# Test with invalid URL
python -m app.cli.summary --url "not-a-url"

# Test with unreachable URL
python -m app.cli.summary --url "https://thisurldoesnotexist12345.com"

# Check errors in database
docker exec -i ratatoskr-postgres psql -U ratatoskr_app -d ratatoskr -c \
  "SELECT id, status, input_url
     FROM requests
    WHERE status = 'error'
    ORDER BY created_at DESC
    LIMIT 5;"

Running pytest

# Run all tests
python -m pytest tests/ -v

# Run a single test file (prefer this for faster iteration)
python -m pytest tests/test_url_utils.py -v

# Run with coverage
python -m pytest tests/ --cov=app --cov-report=html

References

• references/bot-commands.md -- Available bot commands and processing flow
• references/e2e-testing.md -- Docker testing and E2E test setup

Key Files

• CLI Runner: app/cli/summary.py
• Message Router: app/adapters/telegram/message_router.py
• Access Controller: app/adapters/telegram/access_controller.py
• URL Handler: app/adapters/telegram/url_handler.py
• Command Dispatcher: app/adapters/telegram/command_dispatcher.py, app/adapters/telegram/command_dispatch/

Important Notes

• CLI runner does NOT send actual Telegram messages
• Database operations work exactly like production
• The summary validation script exercises both the strict provider schema and the tolerant compatibility mapper used by the live workflow
• Use correlation IDs to trace requests across logs and DB
• Test both English and Russian language flows