# LegiScan MCP Server - E2E Sample Queries
These queries mirror the AutoSkeleton analysis pipeline from April 2025.
Use these to verify the MCP server functions correctly.
---
## 1. Session Discovery (Find available sessions)
**Tool:** `legiscan_get_session_list`
```json
{
"state": "CA"
}
```
**Expected:** Returns session objects with `session_id`, `year_start`, `year_end`, `name`
---
## 2. Bill Search by Keyword (Like scorecard extraction)
**Tool:** `legiscan_search`
```json
{
"query": "housing",
"state": "CA",
"year": 2
}
```
**Expected:** Returns `searchresult.summary` with count + paginated bill results with `bill_id`, `relevance`, `title`
---
## 3. Bill Search by Bill Number (Stage 2 workflow)
**Tool:** `legiscan_search`
```json
{
"query": "AB 1234",
"state": "CA",
"year": 1
}
```
**Expected:** Returns matching bills with `bill_id` for detailed lookup
---
## 4. Get Bill Details (Stage 3 enrichment)
**Tool:** `legiscan_get_bill`
```json
{
"bill_id": 1234567
}
```
*Note: Replace with actual bill_id from search results*
**Expected:** Full bill object with:
- `sponsors[]` (with `people_id`)
- `history[]` (actions)
- `votes[]` (with `roll_call_id`)
- `texts[]` (with `doc_id` for text retrieval)
---
## 5. Get Bill Text (Stage 3 text fetch)
**Tool:** `legiscan_get_bill_text`
```json
{
"doc_id": 2345678,
"decode": true
}
```
*Note: Replace with actual doc_id from bill.texts[]*
**Expected:** Returns `doc` (base64 or decoded HTML), `mime`, `doc_id`
---
## 6. Get Roll Call Votes (Stage 4 salience)
**Tool:** `legiscan_get_roll_call`
```json
{
"roll_call_id": 3456789
}
```
*Note: Replace with actual roll_call_id from bill.votes[]*
**Expected:** Returns:
- `yea`, `nay`, `nv`, `absent` counts
- `votes[]` with individual legislator votes (`people_id`, `vote_id`, `vote_text`)
---
## 7. Get Legislator Info (Party break detection)
**Tool:** `legiscan_get_person`
```json
{
"people_id": 12345
}
```
*Note: Replace with actual people_id from sponsors or roll call votes*
**Expected:** Returns `party`, `role`, `district`, `name`, external IDs
---
## 8. Get All Session Legislators
**Tool:** `legiscan_get_session_people`
```json
{
"session_id": 2024
}
```
*Note: Replace with actual session_id from getSessionList*
**Expected:** Returns list of all active legislators with `people_id`, `party`, `role`
---
## 9. Get Legislator's Sponsored Bills
**Tool:** `legiscan_get_sponsored_list`
```json
{
"people_id": 12345
}
```
**Expected:** Returns list of bills sponsored by this legislator across sessions
---
## Complete Workflow Test Sequence
This mirrors the AutoSkeleton pipeline stages:
### Step 1: Find a session
```
legiscan_get_session_list { "state": "PA" }
```
*Extract a `session_id` for Pennsylvania*
### Step 2: Search for a specific bill
```
legiscan_search { "query": "HB 1437", "state": "PA", "year": 2023 }
```
*Get the `bill_id` from results*
### Step 3: Fetch full bill details
```
legiscan_get_bill { "bill_id": <from_step_2> }
```
*Extract `texts[0].doc_id`, `votes[0].roll_call_id`, `sponsors[0].people_id`*
### Step 4: Fetch bill text for summarization
```
legiscan_get_bill_text { "doc_id": <from_step_3>, "decode": true }
```
### Step 5: Fetch roll call for vote analysis
```
legiscan_get_roll_call { "roll_call_id": <from_step_3> }
```
*Analyze `yea`/`nay` margin, individual votes by party*
### Step 6: Fetch legislator party info
```
legiscan_get_person { "people_id": <from_step_3_or_5> }
```
*Compare legislator `party` against roll call majority for party break detection*
---
## Real Data Examples (Use for actual testing)
### California 2024 Session
```json
// 1. Get sessions
{ "tool": "legiscan_get_session_list", "args": { "state": "CA" } }
// 2. Search current housing bills
{ "tool": "legiscan_search", "args": { "query": "housing affordability", "state": "CA", "year": 2 } }
```
### Federal Congress
```json
// Search federal bills
{ "tool": "legiscan_search", "args": { "query": "infrastructure", "state": "US", "year": 2 } }
```
