Measured harness ledgerPublic result
Claude Fable 5.1Sekiro Main Menu — Claude Fable 5.1 Max
Recreate a high-end cinematic Sekiro-style main menu in the browser from the pinned reference and task contract.
Max reasoningHeadline result
- Workflow cost
- $26.06
- Wall-clock
- 70m (rounded provider display) wall-clock
- Processed tokens
- Not recorded
- Record state
- partial_artifact_build_and_rounded_usage_ledger
Public summary
Claude Fable 5.1 Max partial_artifact_build_and_rounded_usage_ledger ledger: 70m (rounded provider display) wall-clock, Not recorded, and $26.06 provider-reported session cost; not independently recomputable from rounded display values.
Run identity and stack
- Result ID: sekiro-main-menu-browser-fable-5.1-max
- Technical model: claude-fable-5.1
- Provider: Anthropic Claude Code
- Stack: Anthropic Claude Code
- Stack: Technical model/configuration: claude-fable-5.1
- Stack: Browser recreation
- Stack: Pinned visual reference
- Stack: Harness v1 Sekiro prompt
Cost basis
- No raw session export, exact token counts, cache TTL, or per-line price receipt was supplied.
Primary artifact integrity
- Kind: interactive-browser-title-menu
- Path: artifacts/sekiro-main-menu-browser-fable-5.1-max/source/index.html
- SHA-256: e01aec65526763bde15549666230cfcbbef7a10f84a3fffcb581dc2b3b9aa33d
Recorded caveats
- Wall-clock and API duration are whole-minute values displayed by the provider, not raw timestamps.
- The displayed input and cache values are abbreviated; the recorded 61,313,398 processed-token total is an approximation from those displayed components.
- The $26.06 total is provider-reported and is not a verified API-equivalent cost calculation.
- The supplied report does not establish one-shot protocol compliance or session-specific assistant/model-call counts.
- Rolling local-activity requests are excluded because they are not session-attributed benchmark metrics.
- Build/typecheck success does not establish runtime interaction, audio, WebGL visual quality, local FPS, RTX capture, or blind-vote outcomes.
- The supplied usage report identifies a session but does not provide user-turn, assistant-turn, or raw transcript evidence.
Visible evidence gaps
- raw Claude Code session export or exact machine-readable token receipt
- per-line pricing and cache-TTL receipt sufficient to recompute cost
- recorded one-shot protocol evidence
- browser interaction and audio replay receipt
- local FPS with browser, hardware, and viewport receipt
- RTX final capture
- blind-evaluation record
Public result only
This result keeps its public summary and evidence, but it does not currently have a matching Builder test with prompts, projects, Harness workflows or skills.