fixed api calls with seerr, added full context for models, beginning to standardizing single id as source of truths for future tools

.gitignore

@@ -174,3 +174,4 @@ cython_debug/
 # PyPI configuration file
 .pypirc
 
+.docs/

api/ARCHITECTURE.md

@@ -0,0 +1,221 @@
+# API Architecture — Agent + Skill + Tool Pipeline
+
+This document explains how the API routes user messages through the agent/skill/tool pipeline to produce responses.
+
+---
+
+## Overview
+
+```
+┌─────────────────────────────────────────────────────────────────┐
+│                      OpenWebUI / Client                         │
+│  POST /v1/chat/completions  { model, messages, stream }         │
+└──────────────────────────────┬──────────────────────────────────┘
+                               │
+                               ▼
+┌──────────────────────────────────────────────────────────────────┐
+│  api/v1/chat.py  —  chat_completions()                          │
+│                                                                  │
+│  1. _resolve_agent(req.model)  →  Agent                          │
+│  2. agent.build_system_prompt()  →  system prompt                │
+│  3. Build full_messages = [system] + req.messages                │
+│  4. run_agent_with_tools(client, messages, agent_id)             │
+└──────────────────────────────┬───────────────────────────────────┘
+                               │
+                               ▼
+┌──────────────────────────────────────────────────────────────────┐
+│  Tool-Calling Loop  (run_agent_with_tools / run_agent_stream)    │
+│                                                                  │
+│  while turns < max_turns:                                        │
+│    response = LLM.chat(messages, tools=agent_tools)              │
+│    if response has tool_calls:                                   │
+│      for each tool_call:                                         │
+│        result = execute_tool(skills, name, args)                 │
+│        append result to messages                                 │
+│    else:                                                         │
+│      return response.text  (stream tokens if streaming)          │
+└──────────────────────────────────────────────────────────────────┘
+```
+
+---
+
+## Key Concepts
+
+### 1. Agent
+
+An **Agent** is a persona + skill bundle. Defined in `agents/`.
+
+```python
+# agents/media_agent.py
+Agent(
+    agent_id="media-agent",
+    description="Media assistant with Seerr integration",
+    skills=["media_info", "seerr", "triage"],
+    base_prompt="You are a media assistant...",
+)
+```
+
+- `agent_id` — unique name, exposed as a model in OpenWebUI
+- `skills` — list of skill names to load
+- `base_prompt` — starting system prompt, combined with skill fragments
+- `build_system_prompt()` — merges base_prompt + all skill prompt fragments
+
+Agents self-register at import time via `agents/__init__.py`'s `register()`.
+`main.py` calls `load_all_agents()` at startup to import all agent/skill modules.
+
+### 2. Skill
+
+A **Skill** is a capability bundle. Defined in `skills/`.
+
+```python
+# skills/seerr.py
+Skill(
+    name="seerr",
+    description="Seerr integration — trending, discover, request media, submit issues",
+    prompt_fragment="## Seerr Media Tools\n...",
+    tools=[...],          # OpenAI function-calling schema
+    execute=_execute,     # async handler: tool_name + args → ToolResult
+)
+```
+
+- `prompt_fragment` — injected into the agent's system prompt. Teaches the LLM what tools are available and when to use them.
+- `tools` — list of OpenAI function definitions (name, description, parameters).
+- `execute` — async callable that routes tool calls to API handlers.
+
+### 3. Tool
+
+A **Tool** is a single function the LLM can call. Defined as part of a skill's `tools` list.
+
+```python
+{
+    "type": "function",
+    "function": {
+        "name": "seerr_trending",
+        "description": "Get trending movies and TV shows from Seerr...",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "kind": {"type": "string", "enum": ["movie", "tv", "all"]},
+                "language": {"type": "string"},
+            },
+            "required": ["kind"],
+        },
+    },
+}
+```
+
+When the LLM responds with a tool call, the loop:
+1. Extracts `function.name` (e.g. `"seerr_trending"`) and `function.arguments` (e.g. `{"kind": "movie"}`)
+2. Calls `execute_tool(agent.skills, name, args)` which finds the owning skill and runs it
+3. Appends the result text to the message history
+4. Sends back to the LLM for a follow-up response
+
+---
+
+## Full Request Flow
+
+### Step-by-step: "What are trending movies?"
+
+```
+1. OpenWebUI sends:
+   POST /v1/chat/completions
+   {
+     "model": "media-agent",
+     "messages": [
+       {"role": "user", "content": "What are trending movies?"}
+     ],
+     "stream": false
+   }
+
+2. chat_completions():
+   → _resolve_agent(model="media-agent")
+     → get_agent("media-agent") → Agent(skills=["media_info", "seerr", "triage"])
+   → tools = get_all_tools(["media_info", "seerr", "triage"])
+     → Returns 7 tool definitions from seerr.py
+   → system_prompt = agent.build_system_prompt()
+     → base_prompt + media_info fragment + seerr fragment + triage fragment
+
+3. run_agent_with_tools() — Turn 1:
+   → LLM receives: [system prompt with tools] + [user: "What are trending movies?"]
+   → LLM responds: tool_calls = [{"function": {"name": "seerr_trending", "arguments": {"kind": "movie"}}}]
+
+4. Execute tool:
+   → execute_tool(["media_info", "seerr", "triage"], "seerr_trending", {"kind": "movie"})
+   → Finds seerr skill → calls _execute("seerr_trending", ...) → _trending(args)
+   → GET /api/v1/discover/trending?mediaType=movie
+   → Returns formatted list with [tmdb:IDs]
+
+5. run_agent_with_tools() — Turn 2:
+   → LLM receives: previous messages + [tool: "Found 20 trending movies..."]
+   → LLM responds: text = "Here are the top trending movies! 🎬 ..."
+   → finish_reason="stop" → return the text
+
+6. chat_completions() returns:
+   { "choices": [{"message": {"content": "Here are the top trending movies!..."}}] }
+```
+
+### Step-by-step: "Request the 2026 one" (multi-turn context)
+
+```
+1. OpenWebUI sends the FULL history:
+   {
+     "model": "media-agent",
+     "messages": [
+       {"role": "user", "content": "What are trending movies?"},
+       {"role": "assistant", "content": "Here are the top 10 trending movies!
+        1. **Mortal Kombat II** (2026) [tmdb:931285] — ..."},
+       {"role": "user", "content": "could request the mortal kombat one?"},
+       {"role": "assistant", "content": "There are several Mortal Kombat entries! ..."},
+       {"role": "user", "content": "the 2026 one"}
+     ]
+   }
+
+2. chat_completions():
+   → req.messages contains the ENTIRE conversation history
+   → System prompt prepended → full_messages = [system] + 5 history messages
+   → LLM sees everything: the trending list with [tmdb:931285], the disambiguation, "the 2026 one"
+
+3. LLM reasons:
+   - I previously listed Mortal Kombat II (2026) with [tmdb:931285]
+   - The user said "request the mortal kombat one" → I searched and showed 4 options
+   - Now they say "the 2026 one" → that matches Mortal Kombat II (2026) [tmdb:931285]
+   - I should call seerr_request_media(kind="movie", title="Mortal Kombat II", tmdb_id=931285)
+
+4. Tool executes the request → ✅ Success
+```
+
+---
+
+## File Map
+
+```
+main.py                          # FastAPI app entry point, creates singletons
+├── core/
+│   ├── config.py                # .env loader, config constants
+│   └── llm.py                   # create_client() factory for OpenAI client
+├── api/
+│   ├── dependencies.py          # FastAPI Depends: get_llm_client()
+│   └── v1/
+│       └── chat.py              # APIRouter, endpoints, tool-calling loop
+├── agents/
+│   ├── __init__.py              # Agent dataclass, registry, load_all_agents()
+│   ├── naked.py                 # Agent: barebone LLM, no skills
+│   └── media_agent.py           # Agent: media assistant with Seerr skills
+└── skills/
+    ├── __init__.py              # Skill dataclass, ToolResult, registry, execution
+    ├── media_info.py            # Skill: base media assistant persona (prompt-only)
+    ├── seerr.py                 # Skill: Seerr API tools (7 tools, real API calls)
+    └── triage.py                # Skill: fallback for unsupported actions (prompt-only)
+```
+
+## Key Design Decisions
+
+1. **Full multi-turn history**: `req.messages` passes through unchanged. The LLM has access to its own previous responses (including `[tmdb:IDs]`). No external state management needed.
+
+2. **No deterministic pre-processing**: No affirmation detectors, reference resolvers, or hardcoded rules. The LLM interprets user intent naturally from full conversation context.
+
+3. **Agent selection via `model` field**: OpenWebUI sends `model` in the request. `_resolve_agent()` maps it to a registered agent. The `/v1/models` endpoint lists all agents as selectable models.
+
+4. **Skills = prompts + tools**: Skills inject prompt fragments AND optionally expose OpenAI function-calling tools. Prompt-only skills (like `triage`) just shape behavior. Tool-enabled skills (like `seerr`) let the LLM take real actions.
+
+5. **Singleton LLM client**: Created once in `main.py`, stored on `app.state.llm_client`, accessed via FastAPI `Depends(get_llm_client)`.

api/dependencies.py

@@ -3,5 +3,5 @@ from openai import OpenAI
 
 
 def get_llm_client(request: Request) -> OpenAI:
-    """FastAPI dependency - returns the singleton OpenAI client from app.state."""
+    """FastAPI dependency — returns the singleton OpenAI client from app.state."""
     return request.app.state.llm_client

api/v1/chat.py

@@ -7,7 +7,7 @@ import asyncio
 
 from api.dependencies import get_llm_client
 from agents import get as get_agent, list_all as list_all_agents
-from skills import get_all_tools, execute_tool, ToolResult
+from skills import get_all_tools, execute_tool
 
 router = APIRouter()
 
@@ -15,7 +15,7 @@ router = APIRouter()
 class ChatRequest(BaseModel):
     message: str
     session_id: str | None = None
-    agent_id: str | None = None   # which agent to use ("naked", "media-agent", …)
+    agent_id: str | None = None
 
 
 class ChatCompletionRequest(BaseModel):
@@ -30,7 +30,6 @@ class ChatCompletionRequest(BaseModel):
 
 def _resolve_agent(agent_id: str | None = None, model: str | None = None):
     """
-    Resolution order:
     1. explicit agent_id
     2. model field (OpenWebUI sends this — maps to agent_id if registered)
     3. fallback to "naked"
@@ -48,23 +47,18 @@ def _resolve_agent(agent_id: str | None = None, model: str | None = None):
 
 async def run_agent_with_tools(
     client: OpenAI,
-    message: str,
+    messages: list[dict],
     agent_id: str | None = None,
     model: str | None = None,
     max_turns: int = 5,
 ) -> str:
-    """Send the user message to the LLM with tool definitions.
+    """Send messages to the LLM with tool definitions. Tool-calling loop."""
-    Loop: if the LLM responds with tool_calls, execute them and feed
-    results back until the LLM produces a final text answer.
-    """
     agent = _resolve_agent(agent_id, model)
     tools = get_all_tools(agent.skills)
     system_prompt = agent.build_system_prompt()
 
-    messages: list[dict] = [
+    full_messages: list[dict] = [{"role": "system", "content": system_prompt}]
-        {"role": "system", "content": system_prompt},
+    full_messages.extend(messages)
-        {"role": "user", "content": message},
-    ]
 
     loop = asyncio.get_running_loop()
 
@@ -73,129 +67,89 @@ async def run_agent_with_tools(
             None,
             lambda: client.chat.completions.create(
                 model="deepseek-chat",
-                messages=messages,
+                messages=full_messages,
                 tools=tools if tools else None,
                 tool_choice="auto" if tools else None,
             ),
         )
         choice = resp.choices[0]
 
-        # If the model sends a final text answer, return it
         if choice.finish_reason == "stop" and choice.message.content:
             return choice.message.content
 
-        # If the model wants to call tools
         if choice.message.tool_calls:
-            # Append the assistant message with tool_calls
+            full_messages.append(choice.message.model_dump(exclude_none=True))
-            messages.append(choice.message.model_dump(exclude_none=True))
-
             for tc in choice.message.tool_calls:
                 fn_name = tc.function.name
                 fn_args = json.loads(tc.function.arguments)
                 tr = await execute_tool(agent.skills, fn_name, fn_args)
-                result = tr.content if tr else f"Tool '{fn_name}' is not available right now."
+                result = tr.content if tr else f"Tool '{fn_name}' is not available."
-                messages.append({
+                full_messages.append({
-                    "role": "tool",
+                    "role": "tool", "tool_call_id": tc.id, "content": result,
-                    "tool_call_id": tc.id,
-                    "content": result,
                 })
             continue
 
-        # Fallback — should not normally happen
         return choice.message.content or "I'm not sure how to help with that."
 
     return "I've taken several actions but still need more information. Could you clarify?"
 
 
-# ---------------------------------------------------------------------------
-# Non-streaming helper (no tools — used by sync endpoint if tools are absent)
-# ---------------------------------------------------------------------------
-
-def run_agent_simple(
-    client: OpenAI,
-    message: str,
-    agent_id: str | None = None,
-    model: str | None = None,
-) -> str:
-    """Plain LLM call — no tools. Used when the agent has no tool-enabled skills."""
-    agent = _resolve_agent(agent_id, model)
-    response = client.chat.completions.create(
-        model="deepseek-chat",
-        messages=[
-            {"role": "system", "content": agent.build_system_prompt()},
-            {"role": "user", "content": message},
-        ],
-    )
-    return response.choices[0].message.content
-
-
 # ---------------------------------------------------------------------------
 # Streaming generators
 # ---------------------------------------------------------------------------
 
 async def _stream_with_tools(
     client: OpenAI,
-    message: str,
+    messages: list[dict],
     agent_id: str | None = None,
     model: str | None = None,
     max_turns: int = 5,
 ):
-    """Streaming version with tool-calling loop.
+    """Streaming tool-calling loop. Tools run silently, final text is streamed."""
-    Yields tokens from the final text response (tools run silently in the background).
-    """
     agent = _resolve_agent(agent_id, model)
     tools = get_all_tools(agent.skills)
     system_prompt = agent.build_system_prompt()
 
-    messages: list[dict] = [
+    full_messages: list[dict] = [{"role": "system", "content": system_prompt}]
-        {"role": "system", "content": system_prompt},
+    full_messages.extend(messages)
-        {"role": "user", "content": message},
-    ]
 
     loop = asyncio.get_running_loop()
 
     for turn in range(max_turns):
-        # Non-streaming call to check for tool_calls
         resp = await loop.run_in_executor(
             None,
             lambda: client.chat.completions.create(
                 model="deepseek-chat",
-                messages=messages,
+                messages=full_messages,
                 tools=tools if tools else None,
                 tool_choice="auto" if tools else None,
             ),
         )
         choice = resp.choices[0]
 
-        # Tool calls? Execute them and loop
         if choice.message.tool_calls:
-            messages.append(choice.message.model_dump(exclude_none=True))
+            full_messages.append(choice.message.model_dump(exclude_none=True))
             for tc in choice.message.tool_calls:
                 fn_name = tc.function.name
                 fn_args = json.loads(tc.function.arguments)
                 tr = await execute_tool(agent.skills, fn_name, fn_args)
-                result = tr.content if tr else f"Tool '{fn_name}' is not available right now."
+                result = tr.content if tr else f"Tool '{fn_name}' is not available."
-                messages.append({
+                full_messages.append({
                     "role": "tool",
                     "tool_call_id": tc.id,
                     "content": result,
                 })
             continue
 
-        # Final text answer — stream it
         if choice.finish_reason == "stop" and choice.message.content:
-            # Already have a non-streaming answer — yield it token-by-token
             for token in choice.message.content:
                 yield token
                 await asyncio.sleep(0)
             return
 
-        # Last resort: stream the final response
         def _sync_stream():
             stream = client.chat.completions.create(
-                model="deepseek-chat",
+                model="deepseek-chat", messages=full_messages, stream=True,
-                messages=messages,
-                stream=True,
             )
             for chunk in stream:
                 delta = chunk.choices[0].delta
@@ -209,12 +163,12 @@ async def _stream_with_tools(
                 return
             yield token
 
-    yield "…"
+    yield "\u2026"
 
 
 async def run_agent_stream(
     client: OpenAI,
-    message: str,
+    messages: list[dict],
     agent_id: str | None = None,
     model: str | None = None,
 ):
@@ -223,22 +177,20 @@ async def run_agent_stream(
     tools = get_all_tools(agent.skills)
 
     if tools:
-        async for token in _stream_with_tools(client, message, agent_id, model):
+        async for token in _stream_with_tools(client, messages, agent_id, model):
             yield token
         return
 
     # No tools — simple streaming
     system_prompt = agent.build_system_prompt()
+    full_messages: list[dict] = [{"role": "system", "content": system_prompt}]
+    full_messages.extend(messages)
+
     loop = asyncio.get_running_loop()
 
     def _sync_stream():
         stream = client.chat.completions.create(
-            model="deepseek-chat",
+            model="deepseek-chat", messages=full_messages, stream=True,
-            messages=[
-                {"role": "system", "content": system_prompt},
-                {"role": "user", "content": message},
-            ],
-            stream=True,
         )
         for chunk in stream:
             delta = chunk.choices[0].delta
@@ -263,15 +215,17 @@ def root():
 
 
 @router.post("/chat")
-async def chat(req: ChatRequest, client: OpenAI = Depends(get_llm_client)):
+async def chat(
-    """Streaming chat endpoint — returns Server-Sent Events."""
+    req: ChatRequest,
+    client: OpenAI = Depends(get_llm_client),
+):
+    """Streaming chat — single message, no history."""
+    messages = [{"role": "user", "content": req.message}]
+
     async def event_stream():
-        async for token in run_agent_stream(
+        async for token in run_agent_stream(client, messages, req.agent_id):
-            client, req.message, req.agent_id,
-        ):
             payload = json.dumps({"token": token, "session_id": req.session_id})
             yield f"data: {payload}\n\n"
-
         yield f"data: {json.dumps({'done': True, 'session_id': req.session_id})}\n\n"
 
     return StreamingResponse(
@@ -286,24 +240,34 @@ async def chat(req: ChatRequest, client: OpenAI = Depends(get_llm_client)):
 
 
 @router.post("/chat/sync")
-async def chat_sync(req: ChatRequest, client: OpenAI = Depends(get_llm_client)):
+async def chat_sync(
-    """Non-streaming endpoint — uses tool-calling when the agent has tools."""
+    req: ChatRequest,
+    client: OpenAI = Depends(get_llm_client),
+):
+    """Non-streaming chat — single message."""
     agent = _resolve_agent(req.agent_id)
     tools = get_all_tools(agent.skills)
+    messages = [{"role": "user", "content": req.message}]
 
     if tools:
-        response = await run_agent_with_tools(
+        response = await run_agent_with_tools(client, messages, req.agent_id)
-            client, req.message, req.agent_id,
-        )
     else:
-        response = run_agent_simple(client, req.message, req.agent_id)
+        agent_obj = _resolve_agent(req.agent_id)
+        resp = client.chat.completions.create(
+            model="deepseek-chat",
+            messages=[
+                {"role": "system", "content": agent_obj.build_system_prompt()},
+                {"role": "user", "content": req.message},
+            ],
+        )
+        response = resp.choices[0].message.content
 
     return {"response": response, "session_id": req.session_id}
 
 
 @router.get("/agents")
 def list_agents():
-    """Return all registered agents with their ids, descriptions, and skills."""
+    """Return all registered agents."""
     return {
         "agents": [
             {
@@ -318,7 +282,7 @@ def list_agents():
 
 @router.get("/models")
 def list_models():
-    """Return all registered agents as selectable models for OpenWebUI."""
+    """Return agents as selectable models for OpenWebUI."""
     return {
         "object": "list",
         "data": [
@@ -339,36 +303,28 @@ async def chat_completions(
     client: OpenAI = Depends(get_llm_client),
 ):
     """OpenAI-compatible /chat/completions — supports stream=True.
-    Resolves the agent from the model field (OpenWebUI sends this).
+    Multi-turn: req.messages contains the FULL conversation history.
+    Agent resolved from the model field (OpenWebUI sends this).
     """
-    user_message = req.messages[-1]["content"]
     agent = _resolve_agent(model=req.model)
 
     if req.stream:
         async def sse_stream():
-            async for token in run_agent_stream(client, user_message, agent_id=agent.agent_id):
+            async for token in run_agent_stream(
+                client, req.messages, agent_id=agent.agent_id,
+            ):
                 chunk = {
                     "id": "chatcmpl-local",
                     "object": "chat.completion.chunk",
                     "choices": [
-                        {
+                        {"index": 0, "delta": {"content": token}, "finish_reason": None}
-                            "index": 0,
-                            "delta": {"content": token},
-                            "finish_reason": None,
-                        }
                     ],
                 }
                 yield f"data: {json.dumps(chunk)}\n\n"
             final_chunk = {
                 "id": "chatcmpl-local",
                 "object": "chat.completion.chunk",
-                "choices": [
+                "choices": [{"index": 0, "delta": {}, "finish_reason": "stop"}],
-                    {
-                        "index": 0,
-                        "delta": {},
-                        "finish_reason": "stop",
-                    }
-                ],
             }
             yield f"data: {json.dumps(final_chunk)}\n\n"
             yield "data: [DONE]\n\n"
@@ -376,18 +332,23 @@ async def chat_completions(
         return StreamingResponse(
             sse_stream(),
             media_type="text/event-stream",
-            headers={
+            headers={"Cache-Control": "no-cache", "Connection": "keep-alive"},
-                "Cache-Control": "no-cache",
-                "Connection": "keep-alive",
-            },
         )
 
-    # Non-streaming path
+    # Non-streaming — full history, tool-calling
     tools = get_all_tools(agent.skills)
     if tools:
-        response = await run_agent_with_tools(client, user_message, agent_id=agent.agent_id)
+        response = await run_agent_with_tools(
+            client, req.messages, agent_id=agent.agent_id,
+        )
     else:
-        response = run_agent_simple(client, user_message, agent_id=agent.agent_id)
+        system_prompt = agent.build_system_prompt()
+        full_msgs: list[dict] = [{"role": "system", "content": system_prompt}]
+        full_msgs.extend(req.messages)
+        resp = client.chat.completions.create(
+            model="deepseek-chat", messages=full_msgs,
+        )
+        response = resp.choices[0].message.content
 
     return {
         "id": "chatcmpl-local",

skills/seerr.py

@@ -37,7 +37,8 @@ def _client() -> httpx.AsyncClient:
 
 
 def _fmt_items(items: list[dict], kind: str) -> str:
-    """Format a list of media items for the LLM to present."""
+    """Format a list of media items for the LLM to present.
+    Includes the TMDb ID so the LLM can reference it for follow-up actions."""
     lines = []
     for i, item in enumerate(items[:10], 1):
         title = item.get("title") or item.get("name") or "Unknown"
@@ -46,8 +47,10 @@ def _fmt_items(items: list[dict], kind: str) -> str:
             or item.get("firstAirDate", "")[:4]
             or "?"
         )
+        tmdb_id = item.get("id", "")
         overview = (item.get("overview") or "")[:120]
-        lines.append(f"{i}. **{title}** ({year}) — {overview}…")
+        id_tag = f" [tmdb:{tmdb_id}]" if tmdb_id else ""
+        lines.append(f"{i}. **{title}** ({year}){id_tag} — {overview}…")
     return f"Found {len(items)} {kind}. Top results:\n\n" + "\n".join(lines)
 
 
@@ -160,6 +163,104 @@ TOOLS = [
             },
         },
     },
+    {
+        "type": "function",
+        "function": {
+            "name": "seerr_search",
+            "description": "Search for movies, TV shows, or people on Seerr "
+            "by title or name. Uses /search. Call when a user asks 'find me "
+            "the movie X', 'search for show Y', or 'who is actor Z?'.",
+            "parameters": {
+                "type": "object",
+                "properties": {
+                    "query": {
+                        "type": "string",
+                        "description": "The search term — a movie title, "
+                        "TV show name, or person name.",
+                    },
+                    "kind": {
+                        "type": "string",
+                        "enum": ["movie", "tv", "person", "all"],
+                        "description": "Filter by media type. Use 'all' "
+                        "when the user doesn't specify.",
+                    },
+                    "language": {
+                        "type": "string",
+                        "description": "Language filter (e.g. 'en'). "
+                        "Omit for all languages.",
+                    },
+                    "page": {
+                        "type": "integer",
+                        "description": "Page number (default 1).",
+                    },
+                },
+                "required": ["query"],
+            },
+        },
+    },
+    {
+        "type": "function",
+        "function": {
+            "name": "seerr_media_details",
+            "description": "Get full details for a specific movie or TV show "
+            "(cast, crew, runtime, genres, ratings, streaming providers, etc.). "
+            "Call when a user asks 'tell me about movie X' or 'show me details "
+            "for show Y'.",
+            "parameters": {
+                "type": "object",
+                "properties": {
+                    "kind": {
+                        "type": "string",
+                        "enum": ["movie", "tv"],
+                        "description": "Whether to look up a movie or TV show.",
+                    },
+                    "tmdb_id": {
+                        "type": "integer",
+                        "description": "The TMDb ID of the movie or TV show.",
+                    },
+                    "title": {
+                        "type": "string",
+                        "description": "Title to search for if tmdb_id is "
+                        "not known. The system will search and use the first match.",
+                    },
+                    "language": {
+                        "type": "string",
+                        "description": "Language filter (e.g. 'en'). "
+                        "Omit for all languages.",
+                    },
+                },
+                "required": ["kind"],
+            },
+        },
+    },
+    {
+        "type": "function",
+        "function": {
+            "name": "seerr_my_requests",
+            "description": "Get the user's pending, approved, or completed "
+            "media requests from Seerr. Call when a user asks 'what have I "
+            "requested?', 'status of my requests?', or 'did my request go through?'.",
+            "parameters": {
+                "type": "object",
+                "properties": {
+                    "filter": {
+                        "type": "string",
+                        "enum": ["all", "approved", "available", "pending",
+                                 "processing", "unavailable", "failed",
+                                 "deleted", "completed"],
+                        "description": "Filter by request status. "
+                        "Default is 'pending'.",
+                    },
+                    "media_type": {
+                        "type": "string",
+                        "enum": ["movie", "tv", "all"],
+                        "description": "Filter by media type. "
+                        "Default is 'all'.",
+                    },
+                },
+            },
+        },
+    },
     {
         "type": "function",
         "function": {
@@ -218,6 +319,9 @@ async def _execute(tool_name: str, args: dict) -> ToolResult:
         "seerr_discover": _discover,
         "seerr_request_media": _request_media,
         "seerr_submit_issue": _submit_issue,
+        "seerr_search": _search,
+        "seerr_media_details": _media_details,
+        "seerr_my_requests": _my_requests,
     }
     handler = handlers.get(tool_name)
     if not handler:
@@ -277,9 +381,11 @@ async def _trending(args: dict) -> ToolResult:
 async def _discover(args: dict) -> ToolResult:
     """Jellyseerr discover endpoints:
     Genre:  /api/v1/discover/{movies|tv}/genre/{genreId}
-    Studio: /api/v1/discover/{movies|tv}/studio/{studioId}
+    Keyword: /api/v1/search?query=keyword  (free-text search, filtered by mediaType)
-    Keyword: /api/v1/discover/{movies|tv}?query=keyword
+    Studio: NOT SUPPORTED — /discover/studio requires a numeric TMDB studio ID,
-    Language falls back to the base discover endpoint.
+    not a name. Use keyword search as fallback.
+
+    Language is passed to discover or search as appropriate.
     """
     kind = args["kind"]
     genre = args.get("genre", "").strip()
@@ -298,9 +404,8 @@ async def _discover(args: dict) -> ToolResult:
         "war": 10752, "western": 37,
     }
 
-    base = f"/api/v1/discover/{'movies' if kind == 'movie' else 'tv'}"
     params: dict = {"page": page}
-    endpoint = base
+    endpoint: str
 
     if genre:
         genre_id = genre_map.get(genre.lower())
@@ -309,22 +414,52 @@ async def _discover(args: dict) -> ToolResult:
                 f"I don't recognise the genre '{genre}'. "
                 f"Try one of: {', '.join(sorted(genre_map.keys()))}."
             )
-        endpoint = f"{base}/genre/{genre_id}"
+        endpoint = f"/api/v1/discover/{'movies' if kind == 'movie' else 'tv'}/genre/{genre_id}"
+        if language:
+            params["language"] = language
+        async with _client() as c:
+            r = await c.get(endpoint, params=params)
+            r.raise_for_status()
+            results = r.json().get("results", [])
+        desc = genre
     elif studio:
-        endpoint = f"{base}/studio/{studio}"
+        # /discover/studio/{studioId} requires a numeric TMDB studio ID.
+        # Fall back to searching by name via /search.
+        desc = studio
+        search_query = studio
+        endpoint = "/api/v1/search"
+        params["query"] = search_query
+        if language:
+            params["language"] = language
+        async with _client() as c:
+            r = await c.get(endpoint, params=params)
+            r.raise_for_status()
+            results = r.json().get("results", [])
+            # Filter to requested media type
+            results = [item for item in results if item.get("mediaType") == kind]
     elif keyword:
+        # Free-text keyword → use /search, filtered by mediaType
+        desc = keyword
+        endpoint = "/api/v1/search"
         params["query"] = keyword
-        endpoint = base
+        if language:
+            params["language"] = language
+        async with _client() as c:
+            r = await c.get(endpoint, params=params)
+            r.raise_for_status()
+            results = r.json().get("results", [])
+            results = [item for item in results if item.get("mediaType") == kind]
+    else:
+        # Bare discover with no filter
+        endpoint = f"/api/v1/discover/{'movies' if kind == 'movie' else 'tv'}"
+        if language:
+            params["language"] = language
+        async with _client() as c:
+            r = await c.get(endpoint, params=params)
+            r.raise_for_status()
+            results = r.json().get("results", [])
+        desc = kind
 
-    if language:
-        params["language"] = language
-
-    async with _client() as c:
-        r = await c.get(endpoint, params=params)
-        r.raise_for_status()
-        results = r.json().get("results", [])
-
-    desc = genre or studio or keyword or kind
     if not results:
         return ToolResult.ok(f"No {desc} {kind}s found.")
     return ToolResult.ok(_fmt_items(results, f"{desc} {kind}s"))
@@ -335,16 +470,52 @@ async def _request_media(args: dict) -> ToolResult:
     title = args["title"]
     tmdb_id = args.get("tmdb_id")
 
-    # Step 1: Search for the media
     async with _client() as c:
-        r = await c.get("/api/v1/search/", params={"query": quote(title), "page": 1})
+        # --- Fast-path: TMDb ID known — confirm the title and request directly ---
+        if tmdb_id:
+            # Quick lookup to get the correct title for the confirmation message
+            detail_r = await c.get(f"/api/v1/{kind}/{tmdb_id}")
+            if detail_r.status_code == 200:
+                detail = detail_r.json()
+                media_title = detail.get("title") or detail.get("name") or title
+                media_year = (
+                    detail.get("releaseDate", "")[:4]
+                    or detail.get("firstAirDate", "")[:4]
+                    or "?"
+                )
+            else:
+                # Detail lookup failed — fall back to title search
+                pass
+
+            if detail_r.status_code == 200:
+                # Submit directly with the known TMDb ID
+                request_body: dict = {"mediaType": kind, "mediaId": tmdb_id}
+                if kind == "tv":
+                    request_body["seasons"] = "all"
+                req_r = await c.post("/api/v1/request", json=request_body)
+                if req_r.status_code == 201:
+                    return ToolResult.ok(
+                        f"✅ Successfully requested **{media_title}** ({media_year}). "
+                        f"It has been submitted to Seerr and will be processed soon."
+                    )
+                elif req_r.status_code == 409:
+                    return ToolResult.fail(
+                        f"⚠️ **{media_title}** ({media_year}) has already been "
+                        f"requested or is already available."
+                    )
+                else:
+                    return ToolResult.fail(
+                        f"❌ Failed to request **{media_title}** ({media_year}). "
+                        f"Seerr responded with status {req_r.status_code}: {req_r.text[:500]}"
+                    )
+
+        # --- Slow-path: search by title ---
+        r = await c.get("/api/v1/search", params={"query": quote(title), "page": 1})
         r.raise_for_status()
         results = r.json().get("results", [])
 
-        # Filter by mediaType if we have results from unified search
+        # Filter by mediaType (search returns mixed movie/tv/person results)
         filtered = [item for item in results if item.get("mediaType") == kind] if results else []
-        if not filtered:
-            filtered = results  # fallback if mediaType not set
 
         if not filtered:
             return ToolResult.fail(
@@ -352,25 +523,35 @@ async def _request_media(args: dict) -> ToolResult:
                 f"Please double-check the title or provide a TMDb ID."
             )
 
-        # If tmdb_id provided, match it; otherwise use the first result
+        # --- Ambiguity check: more than one match? ---
-        match = None
+        if len(filtered) > 1:
-        if tmdb_id:
+            lines = [
-            match = next(
+                f"⚠️ Multiple matches for \"{title}\". "
-                (item for item in filtered if item.get("id") == tmdb_id),
+                f"Please call `seerr_request_media` again with the "
-                None,
+                f"correct `tmdb_id` and exact title:\n"
-            )
+            ]
-        if not match:
+            for i, item in enumerate(filtered[:10], 1):
-            match = filtered[0]
+                t = item.get("title") or item.get("name", "Unknown")
+                y = (
+                    item.get("releaseDate", "")[:4]
+                    or item.get("firstAirDate", "")[:4]
+                    or "?"
+                )
+                mid = item.get("id", "?")
+                lines.append(
+                    f"{i}. **{t}** ({y}) — `kind=\"{kind}\", "
+                    f"title=\"{t}\", tmdb_id={mid}`"
+                )
+            return ToolResult.ok("\n".join(lines))
 
-        # Seerr's request endpoint expects the local mediaInfo.id
+        # --- Single match — request it ---
-        media_info = match.get("mediaInfo", {})
+        match = filtered[0]
-        media_id = media_info.get("id") or match.get("id")
+        media_id = match.get("id")
         media_title = match.get("title") or match.get("name") or title
         media_year = (
             (match.get("releaseDate") or match.get("firstAirDate") or "?")[:4]
         )
 
-        # Step 2: Submit the request
         request_body = {
             "mediaType": kind,
             "mediaId": media_id,
@@ -398,6 +579,154 @@ async def _request_media(args: dict) -> ToolResult:
             )
 
 
+async def _search(args: dict) -> ToolResult:
+    """Use Jellyseerr's /api/v1/search endpoint.
+    Supports filtering by mediaType (movie | tv | person).
+    """
+    query = args["query"]
+    kind = args.get("kind", "all")
+    language = args.get("language", "").strip() or None
+    page = args.get("page", 1)
+
+    params: dict = {"query": quote(query), "page": page}
+    if language:
+        params["language"] = language
+
+    async with _client() as c:
+        r = await c.get("/api/v1/search", params=params)
+        r.raise_for_status()
+        data = r.json()
+        results = data.get("results", [])
+
+    # Filter by mediaType if requested
+    if kind != "all":
+        results = [item for item in results if item.get("mediaType") == kind]
+
+    label = f"search results for '{query}'"
+    if kind != "all":
+        label += f" ({kind})"
+    if not results:
+        return ToolResult.ok(f"No {label} found.")
+    return ToolResult.ok(_fmt_items(results, label))
+
+
+async def _media_details(args: dict) -> ToolResult:
+    """Fetch full details for a movie or TV show.
+    Resolves the TMDb ID via search if not provided.
+    """
+    kind = args["kind"]
+    tmdb_id = args.get("tmdb_id")
+    title = args.get("title", "").strip()
+    language = args.get("language", "").strip() or None
+
+    params: dict = {}
+    if language:
+        params["language"] = language
+
+    async with _client() as c:
+        # Resolve TMDb ID if needed
+        if not tmdb_id and title:
+            sr = await c.get("/api/v1/search", params={
+                "query": quote(title), "page": 1,
+            })
+            sr.raise_for_status()
+            sresults = sr.json().get("results", [])
+            sresults = [item for item in sresults if item.get("mediaType") == kind]
+            if sresults:
+                tmdb_id = sresults[0].get("id")
+            else:
+                return ToolResult.fail(
+                    f"I couldn't find {kind} '{title}' on Seerr."
+                )
+
+        if not tmdb_id:
+            return ToolResult.fail(
+                "I need either a TMDb ID or a title to look up media details."
+            )
+
+        endpoint = f"/api/v1/{kind}/{tmdb_id}"
+        r = await c.get(endpoint, params=params)
+        r.raise_for_status()
+        data = r.json()
+
+    # Build a concise summary for the LLM
+    title_str = data.get("title") or data.get("name") or "Unknown"
+    year = (
+        data.get("releaseDate", "")[:4]
+        or data.get("firstAirDate", "")[:4]
+        or "?"
+    )
+    overview = data.get("overview", "No overview available.")
+    runtime = data.get("runtime", "?")
+    vote = data.get("voteAverage", "?")
+    genres = ", ".join(g.get("name", "") for g in data.get("genres", [])[:5])
+
+    lines = [
+        f"**{title_str}** ({year}) [tmdb:{tmdb_id}]",
+        f"⭐ {vote}/10  |  ⏱ {runtime} min  |  Genres: {genres or 'N/A'}",
+        f"",
+        f"{overview[:500]}",
+    ]
+
+    # Cast (top 5)
+    cast = (data.get("credits", {}) or {}).get("cast", [])[:5]
+    if cast:
+        lines.append("")
+        lines.append("**Top Cast:** " + ", ".join(
+            c["name"] for c in cast
+        ))
+
+    # Streaming providers
+    providers = data.get("watchProviders", [])
+    if providers:
+        flatrate = []
+        for region in providers:
+            for p in region.get("flatrate", []) or []:
+                if p.get("name") and p["name"] not in flatrate:
+                    flatrate.append(p["name"])
+        if flatrate:
+            lines.append("")
+            lines.append("**Streaming:** " + ", ".join(flatrate[:5]))
+
+    return ToolResult.ok("\n".join(lines))
+
+
+async def _my_requests(args: dict) -> ToolResult:
+    """Fetch the current user's media requests from /request.
+    Filters and sorting are optional.
+    """
+    filter_status = args.get("filter", "pending")
+    media_type = args.get("media_type", "all")
+
+    params: dict = {"filter": filter_status}
+    if media_type != "all":
+        params["mediaType"] = media_type
+
+    async with _client() as c:
+        r = await c.get("/api/v1/request", params=params)
+        r.raise_for_status()
+        data = r.json()
+        results = data.get("results", [])
+
+    if not results:
+        return ToolResult.ok(f"You have no {filter_status} requests right now.")
+
+    lines = []
+    for i, req in enumerate(results[:10], 1):
+        media = req.get("media", {}) or {}
+        title = media.get("title") or media.get("name") or "Unknown"
+        status = req.get("status", "?")
+        status_labels = {1: "Pending", 2: "Approved", 3: "Declined"}
+        status_str = status_labels.get(status, f"Status {status}")
+        is_4k = " (4K)" if req.get("is4k") else ""
+        lines.append(f"{i}. **{title}**{is_4k} — {status_str}")
+
+    total = data.get("pageInfo", {}).get("results", len(results))
+    return ToolResult.ok(
+        f"You have {total} {filter_status} requests:\n\n" + "\n".join(lines)
+    )
+
+
 async def _submit_issue(args: dict) -> ToolResult:
     subject = args["subject"]
     description = args["description"]
@@ -407,23 +736,28 @@ async def _submit_issue(args: dict) -> ToolResult:
 
     body: dict = {
         "issueType": int(issue_type),
-        "subject": subject,
         "message": description,
     }
     if media_title:
         body["message"] = f"[Media: {media_title}]\n\n{description}"
 
     async with _client() as c:
-        # --- Resolve mediaId (Seerr's internal ID, not TMDb) ---
+        # --- Resolve mediaId (Seerr internal ID for /issue endpoint) ---
         if not media_id and media_title:
-            search_r = await c.get("/api/v1/search/", params={"query": quote(media_title), "page": 1})
+            search_r = await c.get("/api/v1/search", params={
-            if search_r.status_code == 200:
+                "query": quote(media_title), "page": 1,
-                results = search_r.json().get("results", [])
+            })
-                if results:
+            search_r.raise_for_status()
-                    # Seerr's /api/v1/issue expects the local mediaInfo.id,
+            results = search_r.json().get("results", [])
-                    # not the TMDb id at the top level.
+
-                    media_info = results[0].get("mediaInfo", {})
+            # Filter to actual media (not persons) and prefer exact title match
-                    media_id = media_info.get("id") or results[0].get("id")
+            media_results = [
+                item for item in results
+                if item.get("mediaType") in ("movie", "tv")
+            ]
+            if media_results:
+                media_info = media_results[0].get("mediaInfo", {})
+                media_id = media_info.get("id") or media_results[0].get("id")
 
         if media_id:
             body["mediaId"] = int(media_id)
@@ -449,18 +783,32 @@ async def _submit_issue(args: dict) -> ToolResult:
 # ---------------------------------------------------------------------------
 seerr_skill = Skill(
     name="seerr",
-    description="Seerr integration — trending, discover, request media, submit issues",
+    description="Seerr integration — search, trending, discover, request media, "
+    "look up details, check requests, submit issues",
     prompt_fragment="""## Seerr Media Tools
 
 You have access to the Seerr media management system. Use the provided tools
 to help users with media-related tasks:
 
+- **seerr_search** — when a user wants to find a specific movie, show, or person
 - **seerr_trending** — when a user asks what is trending/popular/new
 - **seerr_discover** — when a user asks for recommendations by genre/category
+- **seerr_media_details** — when a user wants full info about a movie or show
+- **seerr_my_requests** — when a user asks about their pending/approved requests
 - **seerr_request_media** — when a user wants to request a movie or TV show
 - **seerr_submit_issue** — when a user needs to report a problem or needs an
   operator-only action (like deleting media or cancelling a request)
 
+**TMDb ID Rule**: Every movie and TV show has a unique TMDb ID. When you see
+`[tmdb:123456]` in search/trending/discover results, always **show it to the user**
+in your response. Never strip or omit the TMDb ID when presenting results — the
+user needs it to reference items for follow-up actions. Similarly, capture the ID
+for any follow-up action you take (request details, submit a request, file an
+issue, etc.). If you don't have a TMDb ID and need to take action on a title,
+search first to get one. Never rely on title alone when an ID is available —
+titles are ambiguous, IDs are not. This rule applies to all media tools, present
+and future.
+
 Always confirm successful actions to the user. If a tool fails, tell the user
 what went wrong and suggest alternatives.""",
     tools=TOOLS,