Release 1.1.0

This allows `run-in-container` to default to using the much faster
`build-container-image-debug`.

.github/workflows/workflow.yml

@@ -24,6 +24,10 @@ jobs:
     name: Build and Publish
     runs-on: self-hosted
     steps:
+      - name: Set up QEMU
+        uses: docker/setup-qemu-action@v3
+        with:
+          platforms: arm64
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v1
       - name: Login to ghcr.io

CHANGELOG.md

@@ -1 +1,37 @@
-There's nothing here yet.
+# (2024-09-21) Version 1.1.0
+
+- (**Feature**) Adds support for [prompt variables](./docs/configuration/text-generation.md#️-prompt-override) (date/time, bot name, model id) ([2a5a2d6a4d](https://github.com/etkecc/baibot/commit/2a5a2d6a4dbf5fd7cb504ac07d4187fdc32ae395))
+
+- (**Improvement**) [Dockerfile](./Dockerfile) changes to produce ~20MB smaller container images ([354063abb7](https://github.com/etkecc/baibot/commit/354063abb79035069bd3b26c53214874e9cdd95d))
+
+- (**Improvement**) [Dockerfile](./Dockerfile) changes to optimize local (debug) runs in a container ([c8c5e0e540](https://github.com/etkecc/baibot/commit/c8c5e0e540ab981e849452eb3ddb0378105e1fc6))
+
+- (**Improvement**) CI changes to try and work around multi-arch image issues like [this one](https://github.com/etkecc/baibot/issues/2) ([5de7559ed6](https://github.com/etkecc/baibot/commit/5de7559ed685a41c22dfc12283681f02f4c2ee00))
+
+
+# (2024-09-19) Version 1.0.6
+
+Improvements to:
+
+- messages sent by the bot - better onboarding flow, especially when no agents have been created yet
+- documentation pages
+
+
+# (2024-09-14) Version 1.0.5
+
+Further [improves](https://github.com/etkecc/baibot/commit/3b25b92a81a05ebaf1c6dbabf675fbfbe6c9f418) the typing notification logic, so that it tolerates edge cases better.
+
+
+# (2024-09-14) Version 1.0.4
+
+[Improves](https://github.com/etkecc/baibot/commit/dd1dd78312e3db7f92b37fb3b4750fbe35de7115) the typing notification logic.
+
+
+# (2024-09-13) Version 1.0.3
+
+Contains [fixes](https://github.com/etkecc/rust-mxlink/commit/f339fc85e69aa7f614394ad303d1614cd307319c) for [some](https://github.com/etkecc/baibot/issues/1) startup failures caused by partial initialization (errors during startup).
+
+
+# (2024-09-12) Version 1.0.0
+
+Initial release. 🎉

Cargo.lock

@@ -297,12 +297,13 @@ dependencies = [
 
 [[package]]
 name = "baibot"
-version = "1.0.1"
+version = "1.1.0"
 dependencies = [
  "anthropic-rs",
  "anyhow",
  "async-openai 0.24.0",
  "base64 0.22.1",
+ "chrono",
  "etke_openai_api_rust",
  "matrix-sdk",
  "mxidwc",
@@ -506,6 +507,15 @@ dependencies = [
  "zeroize",
 ]
 
+[[package]]
+name = "chrono"
+version = "0.4.38"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a21f936df1771bf62b77f047b726c4625ff2e8aa607c01ec06e5a05bd8463401"
+dependencies = [
+ "num-traits",
+]
+
 [[package]]
 name = "cipher"
 version = "0.4.4"
@@ -2122,9 +2132,9 @@ dependencies = [
 
 [[package]]
 name = "mxlink"
-version = "1.0.0"
+version = "1.2.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "80cb97e517e3bc8e85b4787ed87037cd324526862ca9375c6d038699010bf6c2"
+checksum = "f1aa513435471af3e1131dc18a1f4eb2e6fb42c11308f1be9b4f0316b3809e32"
 dependencies = [
  "base64 0.22.1",
  "chacha20poly1305",

Cargo.toml

@@ -5,9 +5,9 @@ authors = ["Slavi Pantaleev <slavi@devture.com>"]
 repository = "https://github.com/etkecc/baibot"
 license = "AGPL-3.0-or-later"
 readme = "README.md"
-keywords = ["matrix", "chat", "LLM", "openai", "anthropic", "localai"]
-exclude = [".editorconfig", "justfile", "docs/", "etc/", "Dockerfile", ".dockerignore", ".gitignore", ".github/"]
-version = "1.0.1"
+keywords = ["matrix", "chat", "bot", "AI", "LLM"]
+include = ["/etc/assets/baibot-torso-768.png", "/src", "/README.md", "/CHANGELOG.md", "/LICENSE"]
+version = "1.1.0"
 edition = "2021"
 
 [lib]
@@ -19,10 +19,11 @@ anthropic-rs = "0.1.*"
 anyhow = "1.0.*"
 async-openai = "0.24.*"
 base64 = "0.22.*"
+chrono = { version = "0.4.*", default-features = false, features = ["std", "now"] }
 # We'd rather not depend on this, but we cannot use the ruma-events EventContent macro without it.
 matrix-sdk = { version = "0.7.1", default-features = false }
 mxidwc = "1.0.*"
-mxlink = "1.0.*"
+mxlink = ">=1.2.1"
 etke_openai_api_rust = "0.1.*"
 quick_cache = "0.6.*"
 regex = "1.10.*"

Dockerfile

@@ -4,7 +4,7 @@
 #                                     #
 #######################################
 
-FROM docker.io/rust:1.80.1-slim-bookworm AS build
+FROM docker.io/rust:1.81.0-slim-bookworm AS build
 
 RUN apt-get update && apt-get install -y build-essential pkg-config libssl-dev libsqlite3-dev
 
@@ -15,14 +15,23 @@ WORKDIR /app
 
 COPY . /app
 
+ARG RELEASE_BUILD=true
+
 RUN --mount=type=cache,target=/cargo,sharing=locked \
 	--mount=type=cache,target=/target,sharing=locked \
-	cargo build --release
+	if [ "$RELEASE_BUILD" = "true" ]; then \
+		cargo build --release; \
+	else \
+		cargo build; \
+	fi
 
 # Move it out of the mounted cache, so we can copy it in the next stage.
 RUN --mount=type=cache,target=/target,sharing=locked \
-	cp /target/release/baibot /baibot
-
+	if [ "$RELEASE_BUILD" = "true" ]; then \
+		cp /target/release/baibot /baibot; \
+	else \
+		cp /target/debug/baibot /baibot; \
+	fi
 
 #######################################
 #                                     #
@@ -32,7 +41,9 @@ RUN --mount=type=cache,target=/target,sharing=locked \
 
 FROM docker.io/debian:bookworm-slim
 
-RUN apt-get update && apt-get install -y ca-certificates sqlite3
+RUN apt-get update && apt-get install -y ca-certificates sqlite3 && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
 
 WORKDIR /app
 

docs/access.md

@@ -5,17 +5,21 @@ This bot employs access control to decide who can use its services and manage it
 
 ### 👋 Joining rooms
 
-The bot automatically joins rooms when invited by someone considered a bot [user](#-users).
+The bot automatically joins rooms only when invited by someone considered a bot [👥 user](#-users).
 
 
 ### 👥 Users
 
-The bot will ignore messages (and room invitations) from unallowed users.
-
-Users can **use all the bot's [features](./features.md)** ([💬 Text Generation](./features.md#-text-generation), [🦻 Speech-to-Text](./features.md#-speech-to-text), etc.), but **cannot manage the bot's configuration**.
-
 The bot can be used by users that match some [dynamically](./configuration/README.md#dynamic-configuration) configured [Matrix user id](https://spec.matrix.org/v1.11/#users) patterns.
 
+Users:
+
+- ✅ can **invite the bot to rooms**
+- ✅ can **use all the bot's [features](./features.md)** ([💬 Text Generation](./features.md#-text-generation), [🦻 Speech-to-Text](./features.md#-speech-to-text), etc.) by sending room messages
+- ✅ can **change the bot's configuration in a room** (e.g. `!bai config room ...` commands)
+- ❌ cannot **change the bot's global configuration** (e.g. `!bai config global ...` commands)
+- ❌ cannot **create new [🤖 Agents](./agents.md)** (neither in rooms, nor globally). See [💼 Room-local agent managers](#-room-local-agent-managers) for controlling which users can create agents.
+
 The following commands are available:
 - **Show** the currently allowed users: `!bai access users`
 - **Set** the list of allowed users: `!bai access set-users SPACE_SEPARATED_PATTERNS`
@@ -27,6 +31,8 @@ Example patterns: `@*:example.com @*:another.com @someone:company.org`
 
 Administrators can **manage the bot's configuration and access control**.
 
+Administrators are [👥 Users](#-users) and [💼 Room-local agent managers](#-room-local-agent-managers) implicitly, so they inherit all their permissions.
+
 The bot can be administrated by users that match some [statically](./configuration/README.md#static-configuration) configured [Matrix user id](https://spec.matrix.org/v1.11/#users) patterns.
 
 Administrators cannot be changed without adjusting the bot's configuration on the server.
@@ -35,12 +41,11 @@ Administrators cannot be changed without adjusting the bot's configuration on th
 ### 💼 Room-local agent managers
 
 Room-local agent managers are users privileged to **create their own [agents](./agents.md)** (see `!bai agent`) in rooms.
-Letting regular users create agents which contact arbitrary network services **may be a security issue**.
 
-No room-local agent manager patterns are configured, so new agents can only be created by administrators.
+**⚠️ WARNING**: Letting regular users create agents which contact arbitrary network services **may be a security issue**.
 
 The following commands are available:
 - **Show** the currently allowed users: `!bai access room-local-agent-managers`
 - **Set** the list of allowed users: `!bai access set-room-local-agent-managers SPACE_SEPARATED_PATTERNS`
 
-Example patterns: `@*:synapse.127.0.0.1.nip.io @*:another.com @someone:company.org`
+Example patterns: `@*:example.com @*:another.com @someone:company.org`

docs/configuration/text-generation.md

@@ -68,6 +68,19 @@ Where appropriate, you'll mention best practices and common pitfalls.
 
 A prompt override can also be set globally, see [🛠️ Room Settings](./README.md#room-settings).
 
+Prompts may contain the following **placeholder variables** which will be replaced *every time* the bot is interacted with:
+
+| Placeholder               | Description | Example |
+|---------------------------|-------------|---------|
+| `{{ baibot_name }}`       | Name of the bot as configured in the `user.name` field in the [Static configuration](./README.md#static-configuration) | `Baibot` |
+| `{{ baibot_model_id }}`   | Text-Generation model ID as configured in the [🤖 agent](../agents.md)'s configuration | `gpt-4o` |
+| `{{ baibot_now_utc }}`    | Current date and time in UTC | `2024-09-20 (Friday), 14:26:42 UTC (local timezone/time: unknown)` |
+
+Here's a prompt that combines some of the above variables:
+
+> You are a brief, but helpful bot called {{ baibot_name }} powered by the {{ baibot_model_id }} model. The date/time now is: {{ baibot_now_utc }}."
+
+
 ### 🌡️ Temperature Override
 
 You can override the [temperature](https://blogs.novita.ai/what-are-large-language-model-settings-temperature-top-p-and-max-tokens/#what-is-llm-temperature) (randomness / creativity) parameter configured at the [🤖 agent](../agents.md) level.

docs/installation.md

@@ -15,8 +15,16 @@
 
 We provide prebuilt container images for the `amd64` and `arm64` architectures, so **you don't necessarily need to build images yourself** and can jump to [Running in a container](#-running-in-a-container).
 
-If you nevertheless wish to build a container image yourself, you can do so by running `just build-container-image`.
-This will build and tag your container image as `localhost/baibot:latest`.
+If you nevertheless wish to build a container image yourself, you can do so by running:
+
+- (recommended) `just build-container-image-release` to build a release version of the container image
+
+- or `just build-container-image-debug` to build a debug version of the container image
+
+Debug images are faster to build but are larger in size.
+Release images are ~5x smaller in size, but are slower to build.
+
+Both of these commands will build and tag your container image as `localhost/baibot:latest`.
 
 
 ### 🐋 Running in a container

docs/providers.md

@@ -23,17 +23,20 @@ The list of supported providers is below.
 
 ### How to choose a provider
 
-If you're not sure which provider to start with, we **recommend [OpenAI](#openai)** as it's the most popular and has the **widest range of capabilities**: [💬 text-generation](./features.md#-text-generation), [🖌️ image-generation](./features.md#️-image-generation), [🦻 speech-to-text](./features.md#-speech-to-text), [🗣️ text-to-speech](./features.md#️-text-to-speech).
+If you're not sure which provider to start with, **we recommend [OpenAI](#openai)** as it's the most popular and has the **widest range of capabilities**: [💬 text-generation](./features.md#-text-generation), [🖌️ image-generation](./features.md#️-image-generation), [🦻 speech-to-text](./features.md#-speech-to-text), [🗣️ text-to-speech](./features.md#️-text-to-speech).
 
 You don't need to choose just one though. The bot supports [mixing & matching models](./features.md#-mixing--matching-models), so you can use multiple providers at the same time.
 
 
 ### How to use a provider
 
-- sign up for it
-- obtain an API key
-- [create a new agent](./agents.md#creating-agents)
-- set it as a handler for some types of messages (see [Mixing & matching models](./features.md#-mixing--matching-models)) for a specific room or globally
+1. 📝 **Sign up for it**
+
+2. 🔑 **Obtain an API key**
+
+3. 🤖 **Create one or more agents** in a given room or globally. Next to each provider in the [list below](#supported-providers) you'll see **🗲 Quick start** commands, but you may also refer to the [agent creation guide](./agents.md#creating-agents).
+
+4. 🤝 **Set the new agent as a handler** for a given use-purpose like text-generation, image-generation, etc. The agent creation wizard will tell you how, but you may also refer to the [🤝 Handlers](./configuration/handlers.md) guide.
 
 
 ### Supported providers
@@ -49,7 +52,7 @@ You don't need to choose just one though. The bot supports [mixing & matching mo
   - create a room-local agent: `!bai agent create-room-local anthropic my-anthropic-agent`
   - create a global agent: `!bai agent create-global anthropic my-anthropic-agent`
 
-💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](../sample-provider-configs/anthropic.yml).
+💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](./sample-provider-configs/anthropic.yml).
 
 
 ### Groq
@@ -63,7 +66,7 @@ You don't need to choose just one though. The bot supports [mixing & matching mo
   - create a room-local agent: `!bai agent create-room-local groq my-groq-agent`
   - create a global agent: `!bai agent create-global groq my-groq-agent`
 
-💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](../sample-provider-configs/groq.yml).
+💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](./sample-provider-configs/groq.yml).
 
 
 ### LocalAI
@@ -77,7 +80,7 @@ You don't need to choose just one though. The bot supports [mixing & matching mo
   - create a room-local agent: `!bai agent create-room-local localai my-localai-agent`
   - create a global agent: `!bai agent create-global localai my-localai-agent`
 
-💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](../sample-provider-configs/localai.yml).
+💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](./sample-provider-configs/localai.yml).
 
 
 ### Mistral
@@ -91,7 +94,7 @@ You don't need to choose just one though. The bot supports [mixing & matching mo
   - create a room-local agent: `!bai agent create-room-local mistral my-mistral-agent`
   - create a global agent: `!bai agent create-global mistral my-mistral-agent`
 
-💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](../sample-provider-configs/mistral.yml).
+💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](./sample-provider-configs/mistral.yml).
 
 
 ### Ollama
@@ -105,7 +108,7 @@ You don't need to choose just one though. The bot supports [mixing & matching mo
   - create a room-local agent: `!bai agent create-room-local ollama my-ollama-agent`
   - create a global agent: `!bai agent create-global ollama my-ollama-agent`
 
-💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](../sample-provider-configs/ollama.yml).
+💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](./sample-provider-configs/ollama.yml).
 
 
 ### OpenAI
@@ -122,7 +125,7 @@ For services which are not fully compatible with the OpenAI API, consider using
   - create a room-local agent: `!bai agent create-room-local openai my-openai-agent`
   - create a global agent: `!bai agent create-global openai my-openai-agent`
 
-💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](../sample-provider-configs/openai.yml).
+💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](./sample-provider-configs/openai.yml).
 
 
 ### OpenAI Compatible
@@ -139,7 +142,7 @@ This provider is just as featureful as the [OpenAI](#openai) provider, but is mo
   - create a room-local agent: `!bai agent create-room-local openai-compatible my-openai-compatible-agent`
   - create a global agent: `!bai agent create-global openai-compatible my-openai-compatible-agent`
 
-💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](../sample-provider-configs/openai-compatible.yml).
+💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](./sample-provider-configs/openai-compatible.yml).
 
 
 ### OpenRouter
@@ -153,7 +156,7 @@ This provider is just as featureful as the [OpenAI](#openai) provider, but is mo
   - create a room-local agent: `!bai agent create-room-local openrouter my-openrouter-agent`
   - create a global agent: `!bai agent create-global openrouter my-openrouter-agent`
 
-💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](../sample-provider-configs/openrouter.yml).
+💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](./sample-provider-configs/openrouter.yml).
 
 
 ### Together AI
@@ -167,4 +170,4 @@ This provider is just as featureful as the [OpenAI](#openai) provider, but is mo
   - create a room-local agent: `!bai agent create-room-local together-ai my-together-ai-agent`
   - create a global agent: `!bai agent create-global together-ai my-together-ai-agent`
 
-💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](../sample-provider-configs/together-ai.yml).
+💡 When creating an agent, the bot will show you an up-to-date sample configuration for this provider which looks [like this](./sample-provider-configs/together-ai.yml).

docs/sample-provider-configs/anthropic.yml

@@ -2,7 +2,7 @@ base_url: https://api.anthropic.com/v1
 api_key: YOUR_API_KEY_HERE
 text_generation:
   model_id: claude-3-5-sonnet-20240620
-  prompt: You are a brief, but helpful bot.
+  prompt: "You are a brief, but helpful bot called {{ baibot_name }} powered by the {{ baibot_model_id }} model. The date/time now is: {{ baibot_now_utc }}."
   temperature: 1.0
   max_response_tokens: 8192
   max_context_tokens: 204800

docs/sample-provider-configs/groq.yml

@@ -2,7 +2,7 @@ base_url: https://api.groq.com/openai/v1
 api_key: YOUR_API_KEY_HERE
 text_generation:
   model_id: llama3-70b-8192
-  prompt: You are a brief, but helpful bot.
+  prompt: "You are a brief, but helpful bot called {{ baibot_name }} powered by the {{ baibot_model_id }} model. The date/time now is: {{ baibot_now_utc }}."
   temperature: 1.0
   max_response_tokens: 4096
   max_context_tokens: 131072

docs/sample-provider-configs/localai.yml

@@ -2,7 +2,7 @@ base_url: http://my-localai-self-hosted-service:8080/v1
 api_key: YOUR_API_KEY_HERE
 text_generation:
   model_id: gpt-4
-  prompt: You are a brief, but helpful bot.
+  prompt: "You are a brief, but helpful bot called {{ baibot_name }} powered by the {{ baibot_model_id }} model. The date/time now is: {{ baibot_now_utc }}."
   temperature: 1.0
   max_response_tokens: 4096
   max_context_tokens: 128000

docs/sample-provider-configs/mistral.yml

@@ -2,7 +2,7 @@ base_url: https://api.mistral.ai/v1
 api_key: YOUR_API_KEY_HERE
 text_generation:
   model_id: mistral-large-latest
-  prompt: You are a brief, but helpful bot.
+  prompt: "You are a brief, but helpful bot called {{ baibot_name }} powered by the {{ baibot_model_id }} model. The date/time now is: {{ baibot_now_utc }}."
   temperature: 1.0
   max_response_tokens: 4096
   max_context_tokens: 128000

docs/sample-provider-configs/ollama.yml

@@ -2,7 +2,7 @@ base_url: http://my-ollama-self-hosted-service:11434/v1
 api_key: YOUR_API_KEY_HERE
 text_generation:
   model_id: gemma2:2b
-  prompt: You are a brief, but helpful bot.
+  prompt: "You are a brief, but helpful bot called {{ baibot_name }} powered by the {{ baibot_model_id }} model. The date/time now is: {{ baibot_now_utc }}."
   temperature: 1.0
   max_response_tokens: 4096
   max_context_tokens: 128000

docs/sample-provider-configs/openai-compatible.yml

@@ -2,7 +2,7 @@ base_url: ''
 api_key: YOUR_API_KEY_HERE
 text_generation:
   model_id: some-model
-  prompt: You are a brief, but helpful bot.
+  prompt: "You are a brief, but helpful bot called {{ baibot_name }} powered by the {{ baibot_model_id }} model. The date/time now is: {{ baibot_now_utc }}."
   temperature: 1.0
   max_response_tokens: 4096
   max_context_tokens: 128000

docs/sample-provider-configs/openai.yml

@@ -2,7 +2,7 @@ base_url: https://api.openai.com/v1
 api_key: YOUR_API_KEY_HERE
 text_generation:
   model_id: gpt-4o-2024-08-06
-  prompt: You are a brief, but helpful bot.
+  prompt: "You are a brief, but helpful bot called {{ baibot_name }} powered by the {{ baibot_model_id }} model. The date/time now is: {{ baibot_now_utc }}."
   temperature: 1.0
   max_response_tokens: 16384
   max_context_tokens: 128000

docs/sample-provider-configs/openrouter.yml

@@ -2,7 +2,7 @@ base_url: https://openrouter.ai/api/v1
 api_key: YOUR_API_KEY_HERE
 text_generation:
   model_id: mattshumer/reflection-70b:free
-  prompt: You are a brief, but helpful bot.
+  prompt: "You are a brief, but helpful bot called {{ baibot_name }} powered by the {{ baibot_model_id }} model. The date/time now is: {{ baibot_now_utc }}."
   temperature: 1.0
   max_response_tokens: 2048
   max_context_tokens: 8192

docs/sample-provider-configs/together-ai.yml

@@ -2,7 +2,7 @@ base_url: https://api.together.xyz/v1
 api_key: YOUR_API_KEY_HERE
 text_generation:
   model_id: meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo
-  prompt: You are a brief, but helpful bot.
+  prompt: "You are a brief, but helpful bot called {{ baibot_name }} powered by the {{ baibot_model_id }} model. The date/time now is: {{ baibot_now_utc }}."
   temperature: 1.0
   max_response_tokens: 2048
   max_context_tokens: 8192

etc/app/config.yml.dist

@@ -73,7 +73,7 @@ agents:
     #     api_key: ""
     #     text_generation:
     #       model_id: gpt-4o-2024-08-06
-    #       prompt: You are a brief, but helpful bot.
+    #       prompt: "You are a brief, but helpful bot called {{ baibot_name }} powered by the {{ baibot_model_id }} model. The date/time now is: {{ baibot_now_utc }}."
     #       temperature: 1.0
     #       max_response_tokens: 16384
     #       max_context_tokens: 128000
@@ -97,7 +97,7 @@ agents:
     #     api_key: null
     #     text_generation:
     #       model_id: gpt-4
-    #       prompt: You are a brief, but helpful bot.
+    #       prompt: "You are a brief, but helpful bot called {{ baibot_name }} powered by the {{ baibot_model_id }} model. The date/time now is: {{ baibot_now_utc }}."
     #       temperature: 1.0
     #       max_response_tokens: 16384
     #       max_context_tokens: 128000

justfile

@@ -13,7 +13,8 @@ run-locally *extra_args: app-local-prepare
 	BAIBOT_PERSISTENCE_DATA_DIR_PATH={{ justfile_directory() }}/var/app/local/data \
 	cargo run -- {{ extra_args }}
 
-run-in-container *extra_args: app-container-prepare build-container-image
+# Builds and runs the bot in a container
+run-in-container *extra_args: app-container-prepare build-container-image-debug
 	/usr/bin/env docker run \
 	-it \
 	--rm \
@@ -25,7 +26,7 @@ run-in-container *extra_args: app-container-prepare build-container-image
 	--env BAIBOT_PERSISTENCE_DATA_DIR_PATH=/data \
 	--mount type=bind,src={{ justfile_directory() }}/var/app/container/config.yml,dst=/app/config.yml,ro \
 	--mount type=bind,src={{ justfile_directory() }}/var/app/container/data,dst=/data \
-	{{ container_image_name }} {{ extra_args }}
+	{{ container_image_name }}:latest {{ extra_args }}
 
 # Runs tests
 test *extra_args:
@@ -38,9 +39,16 @@ build-debug *extra_args:
 # Builds an optimized release binary (target/release/*)
 build-release *extra_args: (build-debug "--release")
 
-# Builds a container image
-build-container-image tag='latest':
+# Builds a container image (debug mode)
+build-container-image-debug tag='latest': (_build-container-image "false" tag)
+
+# Builds a container image (release mode)
+build-container-image-release tag='latest': (_build-container-image "true" tag)
+
+_build-container-image release_build tag:
 	/usr/bin/env docker build \
+	--build-arg RELEASE_BUILD={{ release_build }} \
+	-f {{ justfile_directory() }}/Dockerfile \
 	-t {{ container_image_name }}:{{ tag }} \
 	.
 

src/agent/mod.rs

@@ -19,3 +19,7 @@ pub use instantiation::Result as AgentInstantiationResult;
 
 pub use provider::{AgentProvider, AgentProviderInfo, ControllerTrait};
 pub use purpose::AgentPurpose;
+
+pub(super) fn default_prompt() -> &'static str {
+    "You are a brief, but helpful bot called {{ baibot_name }} powered by the {{ baibot_model_id }} model. The date/time now is: {{ baibot_now_utc }}."
+}

src/agent/provider/anthropic/config.rs

@@ -2,7 +2,7 @@ use serde::{Deserialize, Serialize};
 
 use anthropic_rs::models::claude::ClaudeModel;
 
-use crate::agent::provider::ConfigTrait;
+use crate::agent::{default_prompt, provider::ConfigTrait};
 
 #[derive(Debug, Clone, Serialize, Deserialize)]
 pub struct Config {
@@ -58,7 +58,7 @@ impl Default for TextGenerationConfig {
     fn default() -> Self {
         Self {
             model_id: default_text_model_id(),
-            prompt: Some("You are a brief, but helpful bot.".to_owned()),
+            prompt: Some(default_prompt().to_owned()),
             temperature: super::super::default_temperature(),
             max_response_tokens: 8192,
             max_context_tokens: 204_800,

src/agent/provider/anthropic/controller.rs

@@ -95,11 +95,12 @@ impl ControllerTrait for Controller {
             ));
         };
 
-        let prompt_text = params
-            .prompt_override
-            .unwrap_or(self.text_generation_prompt().unwrap_or("".to_owned()))
-            .trim()
-            .to_owned();
+        let prompt_text = params.prompt_variables.format(
+            params
+                .prompt_override
+                .unwrap_or(self.text_generation_prompt().unwrap_or("".to_owned()))
+                .trim(),
+        );
 
         let prompt_message = if prompt_text.is_empty() {
             None
@@ -225,20 +226,25 @@ impl ControllerTrait for Controller {
         }
     }
 
-    fn text_generation_prompt(&self) -> Option<String> {
-        let Some(text_generation_config) = &self.config.text_generation else {
-            return None;
-        };
+    fn text_generation_model_id(&self) -> Option<String> {
+        self.config
+            .text_generation
+            .as_ref()
+            .map(|config| config.model_id.to_owned())
+    }
 
-        text_generation_config.prompt.clone()
+    fn text_generation_prompt(&self) -> Option<String> {
+        self.config
+            .text_generation
+            .as_ref()
+            .and_then(|config| config.prompt.clone())
     }
 
     fn text_generation_temperature(&self) -> Option<f32> {
-        let Some(text_generation_config) = &self.config.text_generation else {
-            return None;
-        };
-
-        Some(text_generation_config.temperature)
+        self.config
+            .text_generation
+            .as_ref()
+            .map(|config| config.temperature)
     }
 
     fn text_to_speech_voice(&self) -> Option<String> {

src/agent/provider/controller.rs

@@ -13,6 +13,8 @@ pub trait ControllerTrait {
 
     fn ping(&self) -> impl std::future::Future<Output = anyhow::Result<PingResult>> + Send;
 
+    fn text_generation_model_id(&self) -> Option<String>;
+
     fn text_generation_prompt(&self) -> Option<String>;
 
     fn text_generation_temperature(&self) -> Option<f32>;
@@ -63,6 +65,14 @@ impl ControllerTrait for ControllerType {
         }
     }
 
+    fn text_generation_model_id(&self) -> Option<String> {
+        match &self {
+            ControllerType::OpenAI(controller) => controller.text_generation_model_id(),
+            ControllerType::OpenAICompat(controller) => controller.text_generation_model_id(),
+            ControllerType::Anthropic(controller) => controller.text_generation_model_id(),
+        }
+    }
+
     fn text_generation_prompt(&self) -> Option<String> {
         match &self {
             ControllerType::OpenAI(controller) => controller.text_generation_prompt(),

src/agent/provider/entity/mod.rs

@@ -9,5 +9,7 @@ pub use agent_provider::{AgentProvider, AgentProviderInfo};
 pub use image_generation::{ImageGenerationParams, ImageGenerationResult};
 pub use ping::PingResult;
 pub use speech_to_text::{SpeechToTextParams, SpeechToTextResult};
-pub use text_generation::{TextGenerationParams, TextGenerationResult};
+pub use text_generation::{
+    TextGenerationParams, TextGenerationPromptVariables, TextGenerationResult,
+};
 pub use text_to_speech::{TextToSpeechParams, TextToSpeechResult};

src/agent/provider/entity/text_generation/mod.rs

@@ -1,8 +1,13 @@
+mod prompt_variables;
+
+pub use prompt_variables::TextGenerationPromptVariables;
+
 #[derive(Default)]
 pub struct TextGenerationParams {
     pub context_management_enabled: bool,
     pub prompt_override: Option<String>,
     pub temperature_override: Option<f32>,
+    pub prompt_variables: TextGenerationPromptVariables,
 }
 
 pub struct TextGenerationResult {

src/agent/provider/entity/text_generation/prompt_variables.rs

@@ -0,0 +1,75 @@
+use chrono::{DateTime, Utc};
+use std::collections::HashMap;
+
+pub struct TextGenerationPromptVariables {
+    map: HashMap<String, String>,
+}
+
+impl Default for TextGenerationPromptVariables {
+    fn default() -> Self {
+        Self::new("unnamed", "unknown-model", Utc::now())
+    }
+}
+
+impl TextGenerationPromptVariables {
+    pub fn new(bot_name: &str, model_id: &str, utc_time: DateTime<Utc>) -> Self {
+        let mut map = HashMap::new();
+
+        map.insert("baibot_name".to_string(), bot_name.to_string());
+        map.insert("baibot_model_id".to_string(), model_id.to_string());
+        map.insert("baibot_now_utc".to_string(), format_utc_time(utc_time));
+
+        Self { map }
+    }
+
+    pub fn format(&self, text: &str) -> String {
+        let mut formatted_text = text.to_string();
+
+        for (key, value) in &self.map {
+            let placeholder = format!("{{{{ {} }}}}", key);
+            formatted_text = formatted_text.replace(&placeholder, value);
+        }
+
+        formatted_text
+    }
+}
+
+fn format_utc_time(time: DateTime<Utc>) -> String {
+    time.format("%Y-%m-%d (%A), %H:%M:%S UTC").to_string()
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use chrono::{TimeZone, Timelike};
+
+    #[test]
+    fn test_new() {
+        // Intentionally injecting some sub-seconds to ensure formatting would ignore them.
+        let now_utc = Utc
+            .with_ymd_and_hms(2024, 9, 20, 18, 34, 15)
+            .unwrap()
+            .with_nanosecond(250000000)
+            .unwrap();
+
+        let variables = TextGenerationPromptVariables::new("baibot", "gpt-4o", now_utc);
+
+        assert_eq!(
+            variables.map.get("baibot_name"),
+            Some(&"baibot".to_string())
+        );
+        assert_eq!(
+            variables.map.get("baibot_model_id"),
+            Some(&"gpt-4o".to_string())
+        );
+        assert_eq!(
+            variables.map.get("baibot_now_utc"),
+            Some(&format_utc_time(now_utc))
+        );
+
+        let prompt = "Hello, I'm {{ baibot_name }} using {{ baibot_model_id }}. The date/time now is {{ baibot_now_utc }}.";
+        let expected = "Hello, I'm baibot using gpt-4o. The date/time now is 2024-09-20 (Friday), 18:34:15 UTC.";
+
+        assert_eq!(variables.format(prompt), expected);
+    }
+}

src/agent/provider/localai/mod.rs

@@ -1,6 +1,5 @@
-// LocalAI is based on OpenAI (async-openai), because it seems to be fully compatible.
-// Moreover, openai_api_rust does not support speech-to-text, so if we wish to use this feature
-// we need to stick to async-openai.
+// At the time of testing, LocalAI can be powered by `openai`, but we use `openai_compat` for better reliability
+// in the event of future updates to `async-openai`.
 
 use super::openai_compat::Config;
 

src/agent/provider/mod.rs

@@ -21,5 +21,5 @@ pub use config::ConfigTrait;
 
 pub use entity::{
     AgentProvider, AgentProviderInfo, ImageGenerationParams, PingResult, SpeechToTextParams,
-    SpeechToTextResult, TextGenerationParams, TextToSpeechParams,
+    SpeechToTextResult, TextGenerationParams, TextGenerationPromptVariables, TextToSpeechParams,
 };

src/agent/provider/openai/config.rs

@@ -1,6 +1,6 @@
 use serde::{Deserialize, Serialize};
 
-use crate::agent::provider::ConfigTrait;
+use crate::agent::{default_prompt, provider::ConfigTrait};
 
 #[derive(Debug, Clone, Serialize, Deserialize)]
 pub struct Config {
@@ -66,7 +66,7 @@ impl Default for TextGenerationConfig {
     fn default() -> Self {
         Self {
             model_id: default_text_model_id(),
-            prompt: Some("You are a brief, but helpful bot.".to_owned()),
+            prompt: Some(default_prompt().to_owned()),
             temperature: super::super::default_temperature(),
             max_response_tokens: 16_384,
             max_context_tokens: 128_000,

src/agent/provider/openai/controller.rs

@@ -86,11 +86,12 @@ impl ControllerTrait for Controller {
             ));
         };
 
-        let prompt_text = params
-            .prompt_override
-            .unwrap_or(self.text_generation_prompt().unwrap_or("".to_owned()))
-            .trim()
-            .to_owned();
+        let prompt_text = params.prompt_variables.format(
+            params
+                .prompt_override
+                .unwrap_or(self.text_generation_prompt().unwrap_or("".to_owned()))
+                .trim(),
+        );
 
         let prompt_message = if prompt_text.is_empty() {
             None
@@ -391,20 +392,25 @@ impl ControllerTrait for Controller {
         }
     }
 
-    fn text_generation_prompt(&self) -> Option<String> {
-        let Some(text_generation_config) = &self.config.text_generation else {
-            return None;
-        };
+    fn text_generation_model_id(&self) -> Option<String> {
+        self.config
+            .text_generation
+            .as_ref()
+            .map(|config| config.model_id.to_owned())
+    }
 
-        text_generation_config.prompt.clone()
+    fn text_generation_prompt(&self) -> Option<String> {
+        self.config
+            .text_generation
+            .as_ref()
+            .and_then(|config| config.prompt.clone())
     }
 
     fn text_generation_temperature(&self) -> Option<f32> {
-        let Some(text_generation_config) = &self.config.text_generation else {
-            return None;
-        };
-
-        Some(text_generation_config.temperature)
+        self.config
+            .text_generation
+            .as_ref()
+            .map(|config| config.temperature)
     }
 
     fn text_to_speech_voice(&self) -> Option<String> {

src/agent/provider/openai_compat/config.rs

@@ -1,5 +1,6 @@
 use serde::{Deserialize, Serialize};
 
+use crate::agent::default_prompt;
 use crate::agent::provider::openai::{
     ImageGenerationConfig as OpenAIImageGenerationConfig,
     SpeechToTextConfig as OpenAISpeechToTextConfig,
@@ -75,7 +76,7 @@ impl Default for TextGenerationConfig {
     fn default() -> Self {
         Self {
             model_id: default_text_model_id(),
-            prompt: Some("You are a brief, but helpful bot.".to_owned()),
+            prompt: Some(default_prompt().to_owned()),
             temperature: super::super::default_temperature(),
             max_response_tokens: 4096,
             max_context_tokens: 128_000,

src/agent/provider/openai_compat/controller.rs

@@ -84,11 +84,12 @@ impl ControllerTrait for Controller {
             ));
         };
 
-        let prompt_text = params
-            .prompt_override
-            .unwrap_or(self.text_generation_prompt().unwrap_or("".to_owned()))
-            .trim()
-            .to_owned();
+        let prompt_text = params.prompt_variables.format(
+            params
+                .prompt_override
+                .unwrap_or(self.text_generation_prompt().unwrap_or("".to_owned()))
+                .trim(),
+        );
 
         let prompt_message = if prompt_text.is_empty() {
             None
@@ -409,20 +410,25 @@ impl ControllerTrait for Controller {
         }
     }
 
-    fn text_generation_prompt(&self) -> Option<String> {
-        let Some(text_generation_config) = &self.config.text_generation else {
-            return None;
-        };
+    fn text_generation_model_id(&self) -> Option<String> {
+        self.config
+            .text_generation
+            .as_ref()
+            .map(|config| config.model_id.to_owned())
+    }
 
-        text_generation_config.prompt.clone()
+    fn text_generation_prompt(&self) -> Option<String> {
+        self.config
+            .text_generation
+            .as_ref()
+            .and_then(|config| config.prompt.clone())
     }
 
     fn text_generation_temperature(&self) -> Option<f32> {
-        let Some(text_generation_config) = &self.config.text_generation else {
-            return None;
-        };
-
-        Some(text_generation_config.temperature)
+        self.config
+            .text_generation
+            .as_ref()
+            .map(|config| config.temperature)
     }
 
     fn text_to_speech_voice(&self) -> Option<String> {

src/bot/implementation.rs

@@ -9,6 +9,7 @@ use mxlink::matrix_sdk::Room;
 
 use mxlink::{
     InitConfig, LoginConfig, LoginCredentials, LoginEncryption, MatrixLink, PersistenceConfig,
+    TypingNoticeGuard,
 };
 
 use mxlink::helpers::account_data_config::{
@@ -210,6 +211,14 @@ impl Bot {
             .await
     }
 
+    pub(crate) async fn start_typing_notice(&self, room: &Room) -> TypingNoticeGuard {
+        self.inner
+            .matrix_link
+            .rooms()
+            .start_typing_notice(room)
+            .await
+    }
+
     pub async fn start(&self) -> anyhow::Result<()> {
         self.rooms().attach_event_handlers().await;
         self.messaging().attach_event_handlers().await;

src/controller/access/dispatching.rs

@@ -13,7 +13,7 @@ pub async fn dispatch_controller(
     match handler {
         AccessControllerType::Help => {}
         _ => {
-            if !message_context.sender_can_manage_global_config()? {
+            if !message_context.sender_can_manage_global_config() {
                 bot.messaging()
                     .send_error_markdown_no_fail(
                         message_context.room(),

src/controller/access/help.rs

@@ -80,24 +80,21 @@ fn build_section_users(
         message.push_str(&strings::access::users_no_patterns());
     }
 
-    let can_manage_global_config = message_context.sender_can_manage_global_config();
-    if let Ok(can_manage_global_config) = can_manage_global_config {
-        if can_manage_global_config {
-            message.push_str("\n\n");
+    if message_context.sender_can_manage_global_config() {
+        message.push_str("\n\n");
 
-            message.push_str(strings::the_following_commands_are_available());
-            message.push('\n');
+        message.push_str(strings::the_following_commands_are_available());
+        message.push('\n');
 
-            message.push_str(&strings::help::access::users_command_get(command_prefix));
-            message.push('\n');
+        message.push_str(&strings::help::access::users_command_get(command_prefix));
+        message.push('\n');
 
-            message.push_str(&strings::help::access::users_command_set(command_prefix));
-            message.push_str("\n\n");
+        message.push_str(&strings::help::access::users_command_set(command_prefix));
+        message.push_str("\n\n");
 
-            message.push_str(&strings::help::access::example_user_patterns(
-                homeserver_name,
-            ));
-        }
+        message.push_str(&strings::help::access::example_user_patterns(
+            homeserver_name,
+        ));
     }
 
     message
@@ -156,27 +153,24 @@ fn build_section_room_local_agent_managers(
         message.push_str(&strings::access::room_local_agent_managers_no_patterns());
     }
 
-    let can_manage_global_config = message_context.sender_can_manage_global_config();
-    if let Ok(can_manage_global_config) = can_manage_global_config {
-        if can_manage_global_config {
-            message.push_str("\n\n");
-            message.push_str(strings::the_following_commands_are_available());
-            message.push('\n');
+    if message_context.sender_can_manage_global_config() {
+        message.push_str("\n\n");
+        message.push_str(strings::the_following_commands_are_available());
+        message.push('\n');
 
-            message.push_str(
-                &strings::help::access::room_local_agent_managers_command_get(command_prefix),
-            );
-            message.push('\n');
+        message.push_str(
+            &strings::help::access::room_local_agent_managers_command_get(command_prefix),
+        );
+        message.push('\n');
 
-            message.push_str(
-                &strings::help::access::room_local_agent_managers_command_set(command_prefix),
-            );
-            message.push_str("\n\n");
+        message.push_str(
+            &strings::help::access::room_local_agent_managers_command_set(command_prefix),
+        );
+        message.push_str("\n\n");
 
-            message.push_str(&strings::help::access::example_user_patterns(
-                homeserver_name,
-            ));
-        }
+        message.push_str(&strings::help::access::example_user_patterns(
+            homeserver_name,
+        ));
     }
 
     message

src/controller/agent/create/mod.rs

@@ -100,8 +100,6 @@ pub async fn handle_room_local(
         return Ok(());
     };
 
-    message_context.room().typing_notice(true).await?;
-
     if !try_to_ping_agent_or_complain(bot, message_context, &parsed_config.agent).await {
         return Ok(());
     }
@@ -140,7 +138,7 @@ pub async fn handle_global(
     provider: &str,
     agent_id_prefixless: &str,
 ) -> anyhow::Result<()> {
-    if !message_context.sender_can_manage_global_config()? {
+    if !message_context.sender_can_manage_global_config() {
         bot.messaging()
             .send_error_markdown_no_fail(
                 message_context.room(),
@@ -215,8 +213,6 @@ pub async fn handle_global(
         return Ok(());
     };
 
-    message_context.room().typing_notice(true).await?;
-
     if !try_to_ping_agent_or_complain(bot, message_context, &parsed_config.agent).await {
         return Ok(());
     }

src/controller/agent/delete/mod.rs

@@ -50,7 +50,7 @@ pub async fn handle(
                 .await
         }
         PublicIdentifier::DynamicGlobal(_) => {
-            if !message_context.sender_can_manage_global_config()? {
+            if !message_context.sender_can_manage_global_config() {
                 bot.messaging()
                     .send_error_markdown_no_fail(
                         message_context.room(),

src/controller/agent/details/mod.rs

@@ -47,7 +47,7 @@ pub async fn handle(
             }
         }
         PublicIdentifier::DynamicGlobal(_) => {
-            if !message_context.sender_can_manage_global_config()? {
+            if !message_context.sender_can_manage_global_config() {
                 bot.messaging()
                     .send_error_markdown_no_fail(
                         message_context.room(),

src/controller/agent/help/mod.rs

@@ -12,10 +12,7 @@ pub async fn handle(bot: &Bot, message_context: &MessageContext) -> anyhow::Resu
 
     message.push_str(&format!("## {}", strings::help::agent::heading()));
     message.push_str("\n\n");
-    message.push_str(&strings::help::agent::intro(
-        bot.command_prefix(),
-        can_manage_agents,
-    ));
+    message.push_str(&strings::help::agent::intro(bot.command_prefix()));
     message.push('\n');
     message.push_str(&strings::help::agent::intro_capabilities());
     message.push_str("\n\n");
@@ -39,7 +36,7 @@ pub async fn handle(bot: &Bot, message_context: &MessageContext) -> anyhow::Resu
         ));
         message.push('\n');
 
-        if message_context.sender_can_manage_global_config()? {
+        if message_context.sender_can_manage_global_config() {
             message.push_str(&strings::help::agent::create_agent_global(
                 bot.command_prefix(),
             ));

src/controller/cfg/dispatching/mod.rs

@@ -40,7 +40,7 @@ async fn dispatch_config_related_handler(
     bot: &Bot,
 ) -> anyhow::Result<()> {
     if let SettingsStorageSource::Global = config_type {
-        if !message_context.sender_can_manage_global_config()? {
+        if !message_context.sender_can_manage_global_config() {
             bot.messaging()
                 .send_error_markdown_no_fail(
                     message_context.room(),

src/controller/chat_completion/mod.rs

@@ -4,7 +4,9 @@ use mxlink::{MatrixLink, MessageResponseType};
 
 use tracing::Instrument;
 
-use crate::agent::provider::{SpeechToTextParams, TextGenerationParams};
+use crate::agent::provider::{
+    SpeechToTextParams, TextGenerationParams, TextGenerationPromptVariables,
+};
 use crate::agent::AgentInstance;
 use crate::agent::AgentPurpose;
 use crate::agent::ControllerTrait;
@@ -43,6 +45,8 @@ pub async fn handle(
 ) -> anyhow::Result<()> {
     let mut original_message_is_audio = false;
 
+    let mut _typing_notice_guard: Option<mxlink::TypingNoticeGuard> = None;
+
     let speech_to_text_flow_type = message_context
         .room_config_context()
         .speech_to_text_flow_type();
@@ -58,11 +62,11 @@ pub async fn handle(
                 return Ok(());
             }
             SpeechToTextFlowType::TranscribeAndGenerateText => {
-                tracing::debug!("Will be trascribing and possibly generating text..");
+                tracing::debug!("Will be transcribing and possibly generating text..");
                 MessageResponseType::InThread(message_context.thread_info().clone())
             }
             SpeechToTextFlowType::OnlyTranscribe => {
-                tracing::debug!("Will only be trascribing audio to text..");
+                tracing::debug!("Will only be transcribing audio to text..");
                 if message_context.thread_info().is_thread_root_only() {
                     MessageResponseType::Reply(message_context.thread_info().root_event_id.clone())
                 } else {
@@ -71,6 +75,10 @@ pub async fn handle(
             }
         };
 
+        if _typing_notice_guard.is_none() {
+            _typing_notice_guard = Some(bot.start_typing_notice(message_context.room()).await);
+        }
+
         let Some(speech_to_text_created_event_id_result) =
             handle_stage_speech_to_text(bot, message_context, audio_content, response_type).await
         else {
@@ -96,6 +104,10 @@ pub async fn handle(
         .room_config_context()
         .should_auto_text_generate(original_message_is_audio)
     {
+        if _typing_notice_guard.is_none() {
+            _typing_notice_guard = Some(bot.start_typing_notice(message_context.room()).await);
+        }
+
         let speech_to_text_created_event_id_reaction_event_id =
             if let Some(speech_to_text_created_event_id) = speech_to_text_created_event_id {
                 let reaction_event_response = bot
@@ -213,6 +225,10 @@ pub async fn handle(
 
     match text_to_speech_stage_params {
         Some(TextToSpeechParams::Perform(text_to_speech_eligible_payload, response_type)) => {
+            if _typing_notice_guard.is_none() {
+                _typing_notice_guard = Some(bot.start_typing_notice(message_context.room()).await);
+            }
+
             let _tts_result = generate_and_send_tts_for_message(
                 bot,
                 matrix_link.clone(),
@@ -337,8 +353,6 @@ async fn handle_stage_text_generation(
     )
     .await?;
 
-    _ = message_context.room().typing_notice(true).await;
-
     let prefixes_to_strip = match controller_type {
         ChatCompletionControllerType::ViaText { prefixes_to_strip } => prefixes_to_strip.clone(),
         ChatCompletionControllerType::ViaAudio => vec![],
@@ -393,6 +407,16 @@ async fn handle_stage_text_generation(
 
     let start_time = std::time::Instant::now();
 
+    let controller = agent.controller();
+
+    let prompt_variables = TextGenerationPromptVariables::new(
+        bot.name(),
+        &controller
+            .text_generation_model_id()
+            .unwrap_or("unknown-model".to_owned()),
+        chrono::Utc::now(),
+    );
+
     let params = TextGenerationParams {
         context_management_enabled: message_context
             .room_config_context()
@@ -405,10 +429,11 @@ async fn handle_stage_text_generation(
         temperature_override: message_context
             .room_config_context()
             .text_generation_temperature_override(),
+
+        prompt_variables,
     };
 
-    let result = agent
-        .controller()
+    let result = controller
         .generate_text(conversation, params)
         .instrument(span)
         .await;
@@ -498,8 +523,6 @@ async fn handle_stage_speech_to_text_actual_transcribing(
         .get_media_content(&media_request, true)
         .await?;
 
-    _ = message_context.room().typing_notice(true).await;
-
     let span = tracing::debug_span!(
         "speech_to_text_generation",
         agent_id = agent.identifier().as_string()

src/controller/help/mod.rs

@@ -3,7 +3,7 @@ use mxlink::MessageResponseType;
 use crate::{entity::MessageContext, strings, Bot};
 
 pub async fn handle(bot: &Bot, message_context: &MessageContext) -> anyhow::Result<()> {
-    let sender_can_manage_global_config = message_context.sender_can_manage_global_config()?;
+    let sender_can_manage_global_config = message_context.sender_can_manage_global_config();
     let sender_can_manage_room_local_agents =
         message_context.sender_can_manage_room_local_agents()?;
 
@@ -18,10 +18,7 @@ pub async fn handle(bot: &Bot, message_context: &MessageContext) -> anyhow::Resu
     // Agents
     message.push_str(&format!("## {}", strings::help::agent::heading()));
     message.push_str("\n\n");
-    message.push_str(&strings::help::agent::intro(
-        bot.command_prefix(),
-        sender_can_manage_room_local_agents,
-    ));
+    message.push_str(&strings::help::agent::intro(bot.command_prefix()));
     message.push_str("\n\n");
     message.push_str(&strings::help::agent::intro_handler_relation(
         bot.command_prefix(),

src/controller/image/generation.rs

@@ -56,8 +56,6 @@ pub async fn handle_image(
         original_prompt.to_owned()
     };
 
-    message_context.room().typing_notice(true).await?;
-
     let span = tracing::debug_span!(
         "image_generation",
         agent_id = agent.identifier().as_string()
@@ -139,7 +137,7 @@ pub async fn handle_sticker(
         return Ok(());
     };
 
-    message_context.room().typing_notice(true).await?;
+    let _typing_notice_guard = bot.start_typing_notice(message_context.room()).await;
 
     let span = tracing::debug_span!(
         "sticker_generation",

src/controller/provider/mod.rs

@@ -9,17 +9,8 @@ pub fn determine_controller(_text: &str) -> ControllerType {
 }
 
 pub async fn handle_help(message_context: &MessageContext, bot: &Bot) -> anyhow::Result<()> {
-    if !message_context.sender_can_manage_room_local_agents()? {
-        bot.messaging()
-            .send_error_markdown_no_fail(
-                message_context.room(),
-                &strings::provider::not_allowed(),
-                MessageResponseType::Reply(message_context.thread_info().root_event_id.clone()),
-            )
-            .await;
-
-        return Ok(());
-    }
+    let can_create_global_agents = message_context.sender_can_manage_global_config();
+    let can_create_room_local_agents = message_context.sender_can_manage_room_local_agents()?;
 
     let mut message = String::new();
     message.push_str(&format!("## {}", strings::help::provider::heading()));
@@ -69,18 +60,26 @@ pub async fn handle_help(message_context: &MessageContext, bot: &Bot) -> anyhow:
             &provider_info,
         ));
 
-        message.push_str("- 🗲 Quick start:\n");
-        message.push_str(&format!(
-            "\t- create a room-local agent: `{command_prefix} agent create-room-local {provider_id} my-{provider_id}-agent`",
-            command_prefix = bot.command_prefix(),
-            provider_id = provider.to_static_str(),
-        ));
-        message.push('\n');
-        message.push_str(&format!(
-            "\t- create a global agent: `{command_prefix} agent create-global {provider_id} my-{provider_id}-agent`",
-            command_prefix = bot.command_prefix(),
-            provider_id = provider.to_static_str(),
-        ));
+        // We always show a "Quick start" section (even to unprivileged users),
+        // because we're talking about it in a previous message.
+        message.push_str("- 🗲 Quick start:");
+        if can_create_room_local_agents {
+            message.push_str(&format!(
+                "\n\t- create a room-local agent: `{command_prefix} agent create-room-local {provider_id} my-{provider_id}-agent`",
+                command_prefix = bot.command_prefix(),
+                provider_id = provider.to_static_str(),
+            ));
+        }
+        if can_create_global_agents {
+            message.push_str(&format!(
+                "\n\t- create a global agent: `{command_prefix} agent create-global {provider_id} my-{provider_id}-agent`",
+                command_prefix = bot.command_prefix(),
+                provider_id = provider.to_static_str(),
+            ));
+        }
+        if !can_create_room_local_agents && !can_create_global_agents {
+            message.push_str(" ask an administrator to create an agent for you (you lack permissions to do so yourself)");
+        }
 
         message.push_str("\n\n");
     }

src/controller/reaction/text_to_speech.rs

@@ -52,6 +52,8 @@ pub(super) async fn handle(
         return Ok(());
     };
 
+    let _typing_notice_guard = bot.start_typing_notice(message_context.room()).await;
+
     crate::controller::utils::text_to_speech::generate_and_send_tts_for_message(
         bot,
         matrix_link,

src/controller/utils/text_to_speech.rs

@@ -18,8 +18,6 @@ pub async fn generate_and_send_tts_for_message(
     text_message_event_id: &OwnedEventId,
     text_content: &str,
 ) -> bool {
-    _ = message_context.room().typing_notice(true).await;
-
     let reaction_event_response = bot
         .reacting()
         .react_no_fail(

src/entity/message_context.rs

@@ -70,12 +70,12 @@ impl MessageContext {
         &self.thread_info
     }
 
-    pub fn sender_can_manage_global_config(&self) -> anyhow::Result<bool> {
-        Ok(self.trigger_event_info.sender_is_admin)
+    pub fn sender_can_manage_global_config(&self) -> bool {
+        self.trigger_event_info.sender_is_admin
     }
 
-    pub fn sender_can_manage_room_local_agents(&self) -> anyhow::Result<bool> {
-        Ok(self.sender_can_manage_global_config()?
+    pub fn sender_can_manage_room_local_agents(&self) -> mxidwc::Result<bool> {
+        Ok(self.sender_can_manage_global_config()
             || self.sender_is_allowed_room_local_agent_manager()?)
     }
 
@@ -102,21 +102,8 @@ impl MessageContext {
         combined
     }
 
-    fn sender_is_allowed_room_local_agent_manager(&self) -> anyhow::Result<bool> {
-        match &self
-            .global_config()
-            .access
-            .room_local_agent_manager_patterns
-        {
-            None => Ok(false),
-            Some(patterns) => {
-                let allowed_regexes = mxidwc::parse_patterns_vector(patterns)?;
-
-                Ok(mxidwc::match_user_id(
-                    self.sender_id().as_str(),
-                    &allowed_regexes,
-                ))
-            }
-        }
+    fn sender_is_allowed_room_local_agent_manager(&self) -> mxidwc::Result<bool> {
+        self.room_config_context()
+            .is_user_allowed_room_local_agent_manager(self.sender_id().clone())
     }
 }

src/entity/room_config_context.rs

@@ -1,3 +1,5 @@
+use mxlink::matrix_sdk::ruma::OwnedUserId;
+
 use super::globalconfig::GlobalConfig;
 use super::roomconfig::RoomConfig;
 
@@ -180,4 +182,18 @@ impl RoomConfigContext {
                     .clone()
             })
     }
+
+    pub fn is_user_allowed_room_local_agent_manager(
+        &self,
+        user_id: OwnedUserId,
+    ) -> mxidwc::Result<bool> {
+        match &self.global_config.access.room_local_agent_manager_patterns {
+            None => Ok(false),
+            Some(patterns) => {
+                let allowed_regexes = mxidwc::parse_patterns_vector(patterns)?;
+
+                Ok(mxidwc::match_user_id(user_id.as_str(), &allowed_regexes))
+            }
+        }
+    }
 }

src/strings/agent.rs

@@ -44,7 +44,7 @@ pub fn not_allowed_to_manage_static_agents() -> String {
 
 pub fn configuration_does_not_result_in_a_working_agent(err: anyhow::Error) -> String {
     format!(
-        "The provided configuration does not result in a working agent. The following error was encountered when trying to talk to the agent API:\n```\n{}```",
+        "The provided configuration does not result in a working agent. The following error was encountered when trying to talk to the agent API:\n```\n{}\n```",
         err,
     )
 }

src/strings/help/agent.rs

@@ -2,15 +2,8 @@ pub fn heading() -> String {
     "🤖 Agents".to_owned()
 }
 
-pub fn intro(command_prefix: &str, can_see_providers: bool) -> String {
-    format!(
-        "An agent is an instantiation and configuration of some **☁️ provider**{}.",
-        if can_see_providers {
-            format!(" (see `{command_prefix} provider`)")
-        } else {
-            "".to_owned()
-        }
-    )
+pub fn intro(command_prefix: &str) -> String {
+    format!("An agent is an instantiation and configuration of some **☁️ provider** (see `{command_prefix} provider`).")
 }
 
 pub fn intro_handler_relation(command_prefix: &str) -> String {
@@ -24,7 +17,7 @@ pub fn intro_capabilities() -> String {
 }
 
 pub fn no_permission_to_create_agents() -> &'static str {
-    "You are neither an administrator, nor a room-local agent manager, so **you cannot create new agents by yourself**."
+    "⚠️ You are neither a bot administrator, nor a room-local agent manager, so **you cannot create new agents by yourself**."
 }
 
 pub fn list_agents(command_prefix: &str) -> String {

src/strings/introduction.rs

@@ -23,20 +23,6 @@ fn purposes_intro() -> &'static str {
     "I can typically be used for the following purposes:"
 }
 
-fn no_text_generation_handler_agent() -> String {
-    format!(
-        "There is no configured handler agent which supports {} {}.",
-        AgentPurpose::TextGeneration.emoji(),
-        AgentPurpose::TextGeneration
-    )
-}
-
-fn introduction_outro(command_prefix: &str) -> String {
-    format!(
-        "You may also send a `{command_prefix} help` command message in this room for more information."
-    )
-}
-
 pub async fn create_on_join_introduction(
     name: &str,
     command_prefix: &str,
@@ -111,18 +97,17 @@ pub async fn create_on_join_introduction(
     message.push_str("\n\n");
 
     if got_text_generation_agent {
-        message.push_str(&simply_send_a_message(
+        message.push_str(&make_use_of_me_simply_send_a_message(
             command_prefix,
             room_config_context.text_generation_prefix_requirement_type(),
         ));
     } else {
-        message.push_str(&no_text_generation_handler_agent());
+        message.push_str(&make_use_of_me_agent_creation(
+            command_prefix,
+            room_config_context.text_generation_prefix_requirement_type(),
+        ));
     }
 
-    message.push_str("\n\n");
-
-    message.push_str(&introduction_outro(command_prefix));
-
     message
 }
 
@@ -130,17 +115,85 @@ pub fn create_short_introduction(name: &str) -> String {
     its_me(name)
 }
 
-fn simply_send_a_message(
+fn make_use_of_me_simply_send_a_message(
+    command_prefix: &str,
+    prefix_requirement_type: TextGenerationPrefixRequirementType,
+) -> String {
+    let message = r#"**To make use of me**:
+
+1. 👋 %send_a_message%
+2. 📖 %learn_more%
+"#;
+
+    message
+        .replace("%command_prefix%", command_prefix)
+        .replace(
+            "%send_a_message%",
+            &send_a_text_message(command_prefix, prefix_requirement_type),
+        )
+        .replace(
+            "%learn_more%",
+            &learn_more_from_usage_or_help(command_prefix),
+        )
+}
+
+fn make_use_of_me_agent_creation(
+    command_prefix: &str,
+    prefix_requirement_type: TextGenerationPrefixRequirementType,
+) -> String {
+    let message = r#"**To make use of me**:
+
+1. ☁️ **Choose an agent provider** (e.g. OpenAI, Mistral, etc). Send a `%command_prefix% provider` command to see the list.
+2. 🤖 %create_one_or_more_agents%
+3. 🤝 %set_new_agent_as_handler%
+4. 👋 %send_a_message%
+5. 📖 %learn_more%
+"#;
+
+    message
+        .replace("%command_prefix%", command_prefix)
+        .replace(
+            "%send_a_message%",
+            &send_a_text_message(command_prefix, prefix_requirement_type),
+        )
+        .replace(
+            "%learn_more%",
+            &learn_more_from_usage_or_help(command_prefix),
+        )
+        .replace(
+            "%create_one_or_more_agents%",
+            &create_one_or_more_agents(command_prefix),
+        )
+        .replace(
+            "%set_new_agent_as_handler%",
+            &set_new_agent_as_handler(command_prefix),
+        )
+}
+
+fn send_a_text_message(
     command_prefix: &str,
     prefix_requirement_type: TextGenerationPrefixRequirementType,
 ) -> String {
     match prefix_requirement_type {
         TextGenerationPrefixRequirementType::No => {
-            "**To make use of me, send a message** in this room (e.g. `Hello!`) and see me reply."
-                .to_owned()
+            "**Send a text message** in this room (e.g. `Hello!`) and see me reply.".to_owned()
         }
         TextGenerationPrefixRequirementType::CommandPrefix => {
-            format!("In this room, I'm configured to require the command prefix (`{command_prefix}`) for text messages.\n**To make use of me, send a prefixed text message** (e.g. `{command_prefix} Hello!`) and see me reply.")
+            format!("In this room, I'm configured to require the command prefix (`{command_prefix}`) for text messages. **Send a prefixed text message** (e.g. `{command_prefix} Hello!`) and see me reply.")
         }
     }
 }
+
+fn learn_more_from_usage_or_help(command_prefix: &str) -> String {
+    format!(
+        "**Learn more** by sending a `{command_prefix} usage` or `{command_prefix} help` command."
+    )
+}
+
+pub fn create_one_or_more_agents(command_prefix: &str) -> String {
+    format!("**Create one or more agents** in this room or globally. The provider help message will show you **🗲 Quick start** commands, but you may also send a `{command_prefix} agent` command to see the guide.")
+}
+
+pub fn set_new_agent_as_handler(command_prefix: &str) -> String {
+    format!("**Set the new agent as a handler** for a given use-purpose like text-generation, image-generation, etc. The agent-creation wizard will tell you how, but you may also send a `{command_prefix} config` command to see the guide (in the 🤖 *Handler Agents* section).")
+}

src/strings/provider.rs

@@ -25,10 +25,6 @@ pub fn invalid_configuration_for_provider(
     )
 }
 
-pub fn not_allowed() -> String {
-    "You are not allowed to see the providers list.".to_owned()
-}
-
 pub fn providers_list_intro() -> String {
     "The list of supported providers is below.".to_owned()
 }
@@ -39,7 +35,7 @@ pub fn help_how_to_choose_heading() -> String {
 
 pub fn help_how_to_choose_description(command_prefix: &str) -> String {
     let str = r#"
-If you're not sure which provider to start with, we **recommend OpenAI** as it's the most popular and has the **widest range of capabilities**.
+If you're not sure which provider to start with, **we recommend OpenAI** as it's the most popular and has the **widest range of capabilities**.
 
 You don't need to choose just one though. The bot supports **mixing & matching models** (by setting different handlers for different types of messages - see `%command_prefix% config`), so you can use multiple providers at the same time.
 "#;
@@ -55,13 +51,21 @@ pub fn help_how_to_use_heading() -> String {
 
 pub fn help_how_to_use_description(command_prefix: &str) -> String {
     let str = r#"
-- sign up for it
-- obtain an API key
-- create a new agent (see `%command_prefix% agent`)
-- set the new agent as a handler for some types of messages (see `%command_prefix% config`)
+1. 📝 **Sign up for it**
+2. 🔑 **Obtain an API key**
+3. 🤖 %create_one_or_more_agents%
+4. 🤝 %set_new_agent_as_handler%
 "#;
 
     str.replace("%command_prefix%", command_prefix)
+        .replace(
+            "%create_one_or_more_agents%",
+            &super::introduction::create_one_or_more_agents(command_prefix),
+        )
+        .replace(
+            "%set_new_agent_as_handler%",
+            &super::introduction::set_new_agent_as_handler(command_prefix),
+        )
         .trim()
         .to_owned()
 }

src/strings/usage.rs

@@ -12,7 +12,7 @@ If there's a text-generation handler agent configured (see `%command_prefix% con
 
 Whether the bot responds depends on the **💬 Text Generation / 🗟 Prefix Requirement** setting (see `%command_prefix% config status`).
 Sometimes, a prefix (e.g. `%command_prefix%`) is required in front of messages sent to the room for the bot to respond.
-For multi-user rooms, this setting defaults to "required"
+For multi-user rooms, this setting defaults to "required".
 
 Room messages start a threaded conversation where you can continue back-and-forth communication with the bot.