docs: add v0.33.1 and v0.33.2 changelog, InteractionMode to CONFIG/CLI/TELEGRAM

- CHANGELOG: v0.33.2 (narrator integration complete) and v0.33.1 (InteractionMode)
- CONFIG: add interaction_mode to top-level config example and env var table
- CLI: add --interaction-mode flag to run flags table
- TELEGRAM: update /mode description to reflect interaction_mode

Author: molty3000

docs/CHANGELOG.md

@@ -1,5 +1,41 @@
 # Changelog
 
+## v0.33.2 (2026-05-23) — Narrator Integration Complete
+
+### Telegram Engaging Mode
+- **Instant progress** — sends an immediate "🤔 Looking into that..." message when the agent starts
+- **Live tool narration** — updates the progress message with emoji-rich descriptions on each tool call
+- **Clean chat** — deletes the progress message when the final answer arrives
+
+### Test Coverage
+- InteractionMode config tests: default, `ODEK_INTERACTION_MODE`, CLI override
+- `/mode` command test: verifies interaction_mode documentation
+
+---
+
+## v0.33.1 (2026-05-23) — InteractionMode & Narrator
+
+### New Feature: InteractionMode
+- `interaction_mode` config field: `"engaging"` (default) or `"verbose"`
+- **Engaging mode** — LLM/narrator-powered emoji-rich progress messages instead of raw tool call output
+- **Verbose mode** — traditional raw tool names, args, and results (existing behavior)
+- `ODEK_INTERACTION_MODE` env var and `--interaction-mode` CLI flag
+
+### New Package: `internal/narrate`
+- Template-based tool narration with emojis (📖 Reading, ✏️ Editing, 🔍 Searching, etc.)
+- `narrate.New(enabled)` constructor — zero deps, zero LLM calls
+- 4 tests, offline fallbacks for all built-in tools
+
+### Integration Points
+- CLI (`--interaction-mode` in run, repl, serve, telegram subcommands)
+- ReAct loop (loop.go) — narrator wired into tool execution and thinking phases
+- Renderer — `NarratorMessage()` for terminal output
+- `NewAgent()` — narrator wired based on `InteractionMode`
+- `/mode` command — documents `interaction_mode` options
+- Config default-overlay: unset defaults to `"engaging"`
+
+---
+
 ## v0.33.0 (2026-05-23) — Performance Release
 
 Six performance improvements across the stack, reducing latency per session by **~30-50%**.

docs/CLI.md

@@ -37,6 +37,7 @@
 | `--max-iter <n>` | int | `90` | Max think→act cycles |
 | `--thinking <level>` | string | profile default | Reasoning depth: `enabled`/`disabled`/`low`/`medium`/`high` |
 | `--sandbox` | bool | false | Execute shell commands inside Docker container |
+| `--interaction-mode <mode>` | string | `engaging` | Tool-call rendering: `engaging` (emoji narration) or `verbose` (raw tool output) |
 | `--no-color` | bool | false | Disable colored terminal output |
 | `--prompt-caching` | bool | false | Enable Anthropic/OpenAI/DeepSeek prompt caching markers |
 | `--no-agents` | bool | false | Skip loading AGENTS.md |

docs/CONFIG.md

@@ -27,6 +27,7 @@ Shared across all projects:
   "thinking": "",
   "max_iterations": 90,
   "sandbox": false,
+  "interaction_mode": "engaging",
   "no_color": false,
   "no_agents": false,
   "system": ""
@@ -59,6 +60,7 @@ Every config knob has a `ODEK_*` counterpart:
 | `ODEK_THINKING` | `--thinking` | string |
 | `ODEK_MAX_ITER` | `--max-iter` | int |
 | `ODEK_SANDBOX` | `--sandbox` | bool |
+| `ODEK_INTERACTION_MODE` | `--interaction-mode` | string |
 | `ODEK_NO_COLOR` | `--no-color` | bool |
 | `ODEK_NO_AGENTS` | `--no-agents` | bool |
 | `ODEK_SYSTEM` | `--system` | string |

docs/TELEGRAM.md

@@ -199,7 +199,7 @@ The handler uses `sync.Map` for `TelegramApprover` instances, keyed by `chatID`.
 | `/new` | Reset the current conversation (clear context) |
 | `/stats` | Show session statistics (turn count, model used, etc.) |
 | `/stop` | Cancel a running agent task |
-| `/mode` | Toggle agent modes (sandbox, verbose) |
+| `/mode` | Show current agent modes (interaction_mode, sandbox, skills) |
 | `/restart` | Gracefully restart the bot process |
 | `/plan <description>` | Create a new plan from a natural language description |
 | `/plans` | List all saved plans |