ralph

1## EXECUTE NOW
2 
3**Target: $ARGUMENTS**
4 
5Parse arguments:
6- N (required unless --dry-run): number of tasks to process
7- --parallel: concurrent claim workers (max 5) + cross-connect validation
8- --batch [id]: process only tasks from specific batch
9- --type [type]: process only tasks at a specific phase (extract, create, reflect, reweave, verify, enrich)
10- --dry-run: show what would execute without running
11- --handoff: output structured RALPH HANDOFF block at end (for pipeline chaining)
12 
13### Step 0: Read Vocabulary
14 
15Read `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.
16 
17**START NOW.** Process queue tasks.
18 
19---
20 
21## MANDATORY CONSTRAINT: SUBAGENT SPAWNING IS NOT OPTIONAL
22 
23**You MUST use the Task tool to spawn a subagent for EVERY task. No exceptions.**
24 
25This is not a suggestion. This is not an optimization you can skip for "simple" tasks. The entire architecture depends on fresh context isolation per phase. Executing tasks inline in the lead session:
26- Contaminates context (later tasks run on degraded attention)
27- Skips the handoff protocol (learnings are not captured)
28- Violates the ralph pattern (one phase per context window)
29 
30**If you catch yourself about to execute a task directly instead of spawning a subagent, STOP.** Call the Task tool. Every time. For every task. Including create tasks. Including "simple" tasks.
31…