pipeline

1## EXECUTE NOW
2 
3**Target: $ARGUMENTS**
4 
5Parse immediately:
6- Source file path: the file to process (required)
7- `--handoff`: output RALPH HANDOFF block at end (for chaining)
8- If target is empty: list files in {DOMAIN:inbox}/ and ask which to process
9 
10### Step 0: Read Vocabulary
11 
12Read `ops/derivation-manifest.md` (or fall back to `ops/derivation.md`) for domain vocabulary mapping. All output must use domain-native terms. If neither file exists, use universal terms.
13 
14**START NOW.** Run the full pipeline.
15 
16---
17 
18## Pipeline Overview
19 
20The pipeline chains four phases. Each phase uses skill invocation or /ralph for subagent-based processing. State lives in the queue file — the pipeline is stateless orchestration on top of stateful queue entries.
21 
22```
23Source file
24    |
25    v
26Phase 1: /seed — create extract task, move source to archive
27    |
28    v
29Phase 2: /reduce (via /ralph) — extract claims from source
30    |
31    v
32Phase 3: /ralph (all claims) — create -> reflect -> reweave -> verify
33    |
34    v
35Phase 4: /archive-batch — move task files, generate summary
36    |
37    v
38Complete
39```
40 
41The pipeline is the convenience wrapper. /ralph is the engine. /seed is the entry point.
42 
43---
44 
45## Phase 1: Seed
46 
47Invoke /seed on the target file to create the extract task, check for duplicates, and move the source to its archive folder.
48 
49**How to invoke:**
50 
51Use the Skill tool if available, otherwise execute the /seed workflow directly:
52- Validate source exists
53…