Output files

lpgcast cast --out DIR (or save_cast(result, DIR)) writes:

File	Holds	Same bytes on replay
graph.json	the schema and the graph: graflo's GraFloOutput	yes
groundings.jsonl	one grounding per vertex, edge and property	yes
pending.jsonl	edges waiting for a secondary-identity lookup at write time	yes
episodes.jsonl	one result per episode	yes
quarantine.jsonl	items left out, failed episodes and unmapped entries	yes
manifest.json	versions, settings, counts and token usage	no: usage differs
write.json	what was sent to the database, with --db	—

JSON is written with sorted keys, so two casts compare with diff.

episodes.jsonl

Field	Holds
source_id	the episode
status	ok (nothing left out), partial (some items quarantined) or failed (nothing written)
failure	why a failed episode failed
selection	whole or slice
retrieval	how selection went: hits per lane, seeds fused, gated out, capped and kept, and the schema's size
requests, repairs	model requests, re-asks and repairs included; typed repair rounds
vertices, edges, pending, quarantined, unmapped	counts
repair_failure	why a repair request failed, when one did; the best earlier reply was kept
fixes	deterministic corrections applied without asking the model: coerced, edge_reversed, quote_normalized, grounded_in_context

pending.jsonl

An edge with an endpoint known only by a secondary identity: its edge_id, its source and target (each a vertex type with either an identity or a match on the secondary key) and its properties.

manifest.json

The output format version (2: groundings may point into context), the LPGCast and graflo versions, the schema as name@version, every model setting that reaches the provider, the token ceiling and slice budget, the counts by status, and usage: calls, cache hits, failures, re-asks, and tokens, billed apart from replayed.