Claude Haiku 5.5 (effort default (no --effort flag) · thinking on) on Build a law firm's deadline calendar from trigger events with calendar and court-day counting, mail service and court holidays

AClaude Haiku 5.5default · thinking on · Claude Code 2.1.299 (headless -p) · cloud · 2026-10-24 · by @bobsiopener

Trust 80%Fully autonomoussuccess · first tryCompare

Estimated cost: $0.0052
Wall time: 21s
Human step-ins: 0
Outcome: Pass

Compare with other models on this task →

Exact prompt

# Build a law firm's deadline calendar from trigger events, counting rules, service by mail and court holidays Inputs: `events.csv` (event_id, matter_id, trigger_date, rule_id, service), `rules.json` (each rule's days, direction and counting method) and `court-holidays.json`. Everything is synthetic. `rules.json` and these rules are the record; where `docketing-note.txt` disagrees, the rules win. Rules: - A court day is a date that is not a Saturday, Sunday or court holiday. - calendar counting: start from the trigger date, don't count the trigger day, and count every day; the deadline is the last day counted. If that day is not a court day, move forward to the next court day for an after rule, and back to the previous court day for a before rule. - court_days counting: count only court days (never the trigger day), forward for after rules and backward for before rules; the deadline is the last court day counted. - service mail adds 3 calendar days to an after rule's deadline once it has been computed; if that lands on a day that is not a court day, move forward to the next court day. Service has no effect on before rules. - Each matter's next deadline is its earliest deadline on or after 2026-11-02 (ties: lowest event number); null if it has none. - Write `out/calendar.json`: `{"deadlines": {"<event_id>": "<YYYY-MM-DD>", ...}, "next_by_matter": {"<matter_id>": {"event_id": <id>, "date": "<YYYY-MM-DD>"} or null, ...}}`. ## How this run works - Work on your own: nobody will answer questions in this task, and writing to `ask/` counts against the run. - Write only under `out/`. Create `out/` if it is missing. Inputs are synthetic; don't fetch anything from the network. - Scoring is partial credit: each listed check is pass/fail, and the score is the share passed. Numbers are compared to within 0.005 unless the task says otherwise.

Compute every deadline in a synthetic firm's docket from its trigger events. Calendar-day rules exclude the trigger day and roll forward for after-deadlines but backward for before-deadlines, court-day rules skip weekends and court holidays (including the Thanksgiving pair and Christmas Eve), mail service adds three days to after-deadlines only and rolls again, and each matter's next deadline on or after the review date is picked with ties by event number. A docketing note gets the day-one, roll direction and mail rules wrong. Original benchmark task written for usecaserank.com (not based on a specific public post). Headless, offline run on a reproducible fixture. Each run's notes say how its cost was measured (estimates are marked).

Success criteria: calendar_after before_rules court_days mail_service next_by_matter

Setup notes: Batch 34 case tq02-law-firm-deadline-calendar (offline, synthetic inputs, seed 2891620380). Autonomous case: no stand-in user and no step-ins expected; any file the solver wrote to ask/ is counted as a measured intervention (0) and blocks a pass. Prompt = TASK.md verbatim. Hidden key kept outside the solver's work dir. Scored with the case's scorer (python3 score.py WORKDIR KEYDIR, absolute paths); pass = every check true and no ask/ files. Tools: Read/Write/Edit/Glob/Grep/Bash, web disabled. Single run, no retries. 1-min load average at scoring 0.21 (8 cores). Time-zone results depend on the system time-zone database: tzdata 2026c-0+deb13u1 (Python zoneinfo). Claude Code's own reported cost for claude-haiku-5.5: $0.005249. Score: 5/5. Cost is an ESTIMATE: measured tokens x pricing.json list price (Claude Pro subscription, nothing billed); input_tokens include cache reads and writes. Single run.

Estimated cost List-price estimate; not actually billed.

Metrics

Estimated cost: $0.0052
Wall time: 21s
Human step-ins: 0
Turns: 5

Model usage
Model	Input tok	Output tok	Cost
AClaude Haiku 5.5anthropic/claude-haiku-5.5effort default (no --effort flag) · thinking on	109,799	3,920	$0.0052

Reproduce this run

Use the use case prompt on the use case page.
Same setup: Claude Haiku 5.5 (anthropic/claude-haiku-5.5, effort default (no --effort flag) · thinking on) via Claude Code 2.1.299 (headless -p), cloud.
It counts as a success only if it meets every success criterion: calendar_after before_rules court_days mail_service next_by_matter
Log model, tokens, wall time, human step-ins and the evidence log, then submit it as a reproduction. A different model? Submit a head-to-head rerun.

How runs are scored: Methodology.

Was this useful?

⇄ Rerun on another model

Automated checks

Trust 80% comes from these checks (how Trust % works).

Run date is not in the future : passed
Model is in the dated pricing table : passed
Cost recomputed from tokens matches : passed
Cached tokens are consistent : passed
Output throughput is plausible : passed
Autonomy matches step-ins : passed
Step-ins do not exceed turns : passed
Retries agree with first-try result : passed
Evidence log SHA-256 matches : passed
Evidence log is substantial : passed

Evidence log

[
 {
  "duration_api_ms": 19782,
  "stop_reason": "end_turn",
  "session_id": "[REDACTED]",
  "total_cost_usd": 0.005248640000000001,
  "usage": {
   "input_tokens": 10,
   "cache_creation_input_tokens": 11525,
   "cache_read_input_tokens": 98264,
   "output_tokens": 3920,
   "output_tokens_details": {
    "thinking_tokens": 2009
   },
   "server_tool_use": {
    "web_search_requests": 0,
    "web_fetch_requests": 0
   },
   "service_tier": "standard",
   "cache_creation": {
    "ephemeral_1h_input_tokens": 11525,
    "ephemeral_5m_input_tokens": 0
   },
   "inference_geo": "global",
   "iterations": [
    {
     "input_tokens": 2,
     "output_tokens": 1775,
     "cache_read_input_tokens": 23386,
     "cache_creation_input_tokens": 678,
     "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 678
     },
     "type": "message"
    }
   ],
   "speed": "standard",
   "fallback_credit": null
  },
  "modelUsage": {
   "claude-haiku-5-5": {
    "inputTokens": 10,
    "outputTokens": 3920,
    "cacheReadInputTokens": 98264,
    "cacheCreationInputTokens": 11525,
    "webSearchRequests": 0,
    "costUSD": 0.005248640000000001,
    "contextWindow": 1000000,
    "maxOutputTokens": 128000,
    "thinkingTokens": 2009,
    "canonicalModel": "claude-haiku-5-5",
    "provider": "firstParty",
    "costBasis": "list"
   }
  },
  "permission_denials": [],
  "terminal_reason": "completed",
  "fast_mode_state": "off",
  "fast_mode_disabled_reason": "sdk_opt_in_required",
  "subagent_stats": {
   "spawned": 0,
   "requested": {
    "background": 0,
    "foreground": 0,
    "unset": 0
   },
   "started_in_background": 0,
   "max_depth": 0,
   "spawned_by_subagents": 0,
   "completed": 0,
   "failed": 0,
   "killed": {
    "parent": 0,
    "user": 0,
    "system": 0
   },
   "refused": {
    "depth_limit": 0,
    "concurrency_limit": 0,
    "budget": 0
   },
   "by_type": {}
  },
  "safety_stops": 0,
  "is_error": false,
  "num_turns": 5,
  "subtype": "success",
  "api_error_status": null,
  "result": "I wrote `out/calendar.json` with deadlines for all 13 events and the next deadline for each of the 5 matters.\n\n| Matter | Next event | Date |\n|---|---|---|\n| MAT-2361 | EV-118 | 2026-11-10 |\n| MAT-2593 | EV-112 | 2026-11-09 |\n| MAT-2756 | EV-102 | 2026-11-04 |\n| MAT-2828 | EV-136 | 2027-01-25 |\n| MAT-2914 | EV-126 | 2026-12-16 |\n\nI followed the rules in the task, not `docketing-note.txt`, which conflicts with them on three points:\n- The trigger day is not counted as day one.\n- A weekend deadline moves to the next or previous court day, not always to Monday.\n- Mail service adds 3 days to any after rule, not only appeals. Only EV-112 is affected, since it is the only mail-served after-rule event.\n\nI checked several deadlines by hand against the holiday list and they match. The script is at `/tmp/calc/calc.py`, outside the work directory.\n\nTwo choices to confirm:\n- `next_by_matter` uses the event ID as a string (e.g. `\"EV-118\"`), matching the `deadlines` keys, since the IDs are not plain integers.\n- MAT-2828 has two events due 2027-01-25 (EV-136 and EV-138). I took the lower event number, EV-136, as the tie-break rule says.",
  "ttft_ms": 1606,
  "type": "result",
  "duration_ms": 20134,
  "uuid": "[REDACTED]",
  "ttft_stream_ms": 1425,
  "time_to_request_ms": 204,
  "first_content_frame_ms": 1425,
  "queued_turn_count": 0,
  "result_index": 0
 }
]