derive-budget-allocation-is-accurate
IN derived (depth 1)
The derive pipeline's proportional belief-budget allocation produces correct per-agent token counts
Summary
The system that generates new beliefs from existing ones correctly divides its available token budget across multiple agents based on how many beliefs each agent has, so no agent gets starved of context or wastefully over-allocated. This works because the proportional math is sound and the prompt format round-trips cleanly through serialization, meaning the budget numbers that go in are the ones that actually get used.
Justifications
SL — Budget allocation is correct only when the belief count is correct; the per-belief loop count bug inflates counts and distorts proportional allocation
Antecedents (all must be IN):
- derive-agent-budget-proportional — When agents are present, `_build_beliefs_section` allocates prompt token budget proportionally to each agent's belief count, with a floor of 5 beliefs per agent
- derive-prompt-roundtrips-through-parser — The `### DERIVE` / `### GATE` format is a shared contract between `DERIVE_PROMPT` LLM output, `parse_proposals()` input, and `write_proposals_file()` output, forming a closed serialization loop
Unless (any of these IN defeats this justification):
- derive-agent-count-bug — `_build_beliefs_section` has a bug: `count += len(belief_ids)` is inside the per-belief loop instead of outside it, inflating the count and shrinking the non-agent budget below intended size
Dependents
These beliefs depend on this one:
- derive-prompt-is-deterministic-and-reproducible — The derive pipeline's prompt construction is fully reproducible: deterministic sampling with fixed seeds selects consistent belief subsets, and accurate proportional budget allocation ensures each agent receives the same token share across runs.
- token-budgets-are-accurate-bidirectionally — Token budget management is accurate in both directions: the compact module reliably constrains output size for context-limited consumers, while the derive pipeline correctly allocates input budgets per agent — ensuring resource-bounded operation across the entire LLM integration surface.