NisseBolt App: Missing Functionality vs Slack Bolt Python

This document compares the NisseBolt App class (the Mug plugin interface) against the Slack Bolt Python API and identifies functionality that is missing or incomplete.

Note: some of this functionality already exists in the newer NBServer class (NB/NBServer.h), which is a standalone server rather than a Mug plugin. Where applicable, this is noted below.

3. Shortcuts (shortcut())

Bolt Python:

@app.shortcut("open_modal")
def handle_shortcut(ack, shortcut, client):
    ack()
    client.views_open(trigger_id=shortcut["trigger_id"], view={...})

NisseBolt App: Not supported.

NBServer: Supported via shortcut(std::string callbackId, ShortcutHandler handler).

Impact: Medium. Shortcuts (both global and message shortcuts) allow users to trigger actions from the Slack UI without typing a command. Less commonly used than commands or actions but important for polished integrations.

5. External Select Options (options())

Bolt Python:

@app.options("external_select_action")
def handle_options(ack):
    ack(options=[{"text": {"type": "plain_text", "text": "Option 1"}, "value": "1"}])

NisseBolt App: Not supported.

NBServer: Supported via options(std::string actionId, OptionsHandler handler).

Impact: Low-Medium. Only needed when using external data source select menus in Block Kit.

7. respond() in Handlers

Bolt Python:

@app.command("/echo")
def handle_echo(ack, respond, command):
    ack()
    respond(f"You said: {command['text']}")  # Uses response_url

The respond() function posts an ephemeral or in-channel message using the response_url provided in the interaction payload. Unlike say(), it can send ephemeral messages (visible only to the user who triggered the action).

NisseBolt App: Not supported. Only Say is available, which always uses the chat.postMessage API.

NBServer: Supported via Respond& parameter in command and action handlers.

Impact: Medium. Without respond(), Bots cannot send ephemeral responses to slash commands or interactive actions. All responses via say() are visible to the entire channel.

8. Middleware (use())

Bolt Python:

@app.use
def log_request(body, next):
    print(f"Received: {body['type']}")
    next()

Middleware runs before handlers and can short-circuit the processing chain.

NisseBolt App: Not supported. There is no middleware pipeline.

NBServer: Supported via use(Middleware mw).

Impact: Medium. Middleware is useful for cross-cutting concerns like logging, authorization checks, and rate limiting. Without it, these concerns must be duplicated in every handler.

9. Error Handling (onError())

Bolt Python:

@app.error
def handle_errors(error, body, logger):
    logger.exception(f"Error: {error}")

NisseBolt App: Errors in Say::sendMessage are logged via ThorsLogError but there is no user-configurable error handler. If a handler throws an exception, behavior is undefined at the application level.

NBServer: Supported via onError(ErrorHandler handler).

Impact: Medium. Production bots need a way to handle errors gracefully (alerting, fallback responses, etc.).

Summary

Feature Bolt Python NisseBolt App NBServer Priority
Slash commands Yes No Yes High
Interactive actions Yes No Yes High
Modal views Yes No Yes Medium-High
Shortcuts Yes No Yes Medium
External options Yes No Yes Low-Medium
ack() in handlers Yes Auto only Yes Needed with above
respond() Yes No Yes Medium
Middleware Yes No Yes Medium
Error handler Yes No Yes Medium
Regex match groups Yes No No Low
Event filters String-based Type-based (better) Type-based N/A

The most impactful gaps are slash commands and interactive components (actions + views). These are fundamental to most Slack Bot integrations beyond simple message listeners. The NBServer already implements all of these, so the functionality exists in the codebase but hasn’t been surfaced in the Mug plugin App class.