# Span-level replay: fork from the failure (P12) — TextForAgents LLC
Fork-replay ONE recorded span (a JSON tool or model step from a tracer) out of a failed agent run and get an attribution verdict — determinism / tool-nondeterminism / tool-drift / env-drift / world-state / model-decision — without re-running the whole run. Ingests normalized JSONL span records and OTel-shaped spans (including LangGraph/Langfuse-shaped foreign traces). Side-effecting spans are never re-executed: they are served the recorded result from a fingerprint-keyed stub registry and fail closed on a miss. The recorded env_digest is verified before any execution; a tampered env is refused, never attributed against.
Prices: tool-span replay $0.02 per replay; model-decision replay $0.05 per replay — pay per replay (x402, Base USDC, payTo 0xAc2E243a075fEe442e4c6864DcF880eb4e3A3862), no subscription, NO free tier, NO trial.
Verdicts: determinism / tool-nondeterminism / tool-drift / env-drift / world-state / model-decision (+ UNREPLAYABLE with reason).
Limits: Accuracy is claimed only for the seeded lab set: 24/24 attribution vs 0.167 for the full-re-run baseline (LABREPORT-008). Real-trace sampling is a packet item; no accuracy beyond the seeded set is claimed. Recorder/enricher is a claim limit: traces recorded without tool_version/tool_schema_hash/side_effecting enrichment get reduced verdicts, labeled drift-unavailable, never guessed. Full drift attribution requires the recording contract (env with tool_version + tool_schema_hash and an env_digest).
Start: POST /register for an account token, then POST /mcp (initialize / tools/list / tools/call replay_span) with a verified x402 X-PAYMENT per call.
Support: support@textforagents.com