Spring website traffic by month, crawler taken out
The request
The task request supplied to the adapter, alongside the workspace below. Runtime instructions, authorized tools and skills also affect the system. Containerized runs withhold host-side references; local mode is not an isolation boundary.
Mira wants the website traffic for March, April and May as a monthly summary with a total for the three months. Save it as traffic.xlsx with live formulas, and put what she should know in memo.md. Tom's note explains the Google Analytics files.
Workspace
- ga4_traffic_acquisition_daily.csv
- ga4_users_by_month.csv
- note_from_tom.txt
- search_console_performance_2026-03-01_2026-05-31.csv
Expected deliverables
- memo.md
- traffic.xlsx
Recorded acceptance checks
The table uses check names and outcomes from the frozen campaign ledger. Acceptance requires all required predicates to pass; a predicate may itself contain a tolerance or coverage threshold. Its scope is part of the measurement contract.
| Recorded check | Type | Required | Failed / observed |
|---|---|---|---|
| traffic.xlsx exists | file_exists | yes | 0 / 6 |
| live formulas | xlsx_has_formulas | yes | 0 / 6 |
| no error cells | xlsx_no_errors | yes | 0 / 6 |
| April sessions, crawler removed | xlsx_value_present | yes | 0 / 6 |
| sessions, March to May | xlsx_value_present | yes | 0 / 6 |
| May users (monthly unique, not summed daily) | xlsx_value_present | yes | 0 / 6 |
| memo names the crawl day | text_sentence_matches | yes | 0 / 6 |
| memo carries the crawler sessions and corrected April sessions | text_numbers_present | yes | 0 / 6 |
Current task declaration (8 checks)
From task.yaml. Original runner grading and frozen retrospective scoring are separate. Read the frozen source for the rules behind the reported results.
- traffic.xlsx exists —
file_exists, required - live formulas —
xlsx_has_formulas, required - no error cells —
xlsx_no_errors, required - April sessions, crawler removed —
xlsx_value_present, required - sessions, March to May —
xlsx_value_present, required - May users (monthly unique, not summed daily) —
xlsx_value_present, required - memo names the crawl day —
text_sentence_matches, required - memo carries the crawler sessions and corrected April sessions —
text_numbers_present, required
Recorded results, desk release
| System | Repetitions | Frozen / 3 | Original / 3 | Median minutes | Mean model estimate (USD) |
|---|---|---|---|---|---|
| Proto / DeepSeek V4.1 Flash | 3 | 3 | 7.2 | 0.0801 | |
| Codex / GPT-5.6 sol | 3 | 3 | 6.6 | 0.7413 |
These are repeated executions on the released fixture, not independent new business cases. Cost is captured model usage priced under the recorded assumptions, excluding infrastructure and human review. Per-attempt check verdicts, raw and frozen verdicts, usage, receipt and artifact hashes for this task are in results/latest/attempts.jsonl under "task": "web-traffic-summary".
Recorded results, desk comparison, 28 September 2026
| System | Repetitions | Frozen / 3 | Original / 3 | Median minutes |
|---|---|---|---|---|
| Proto / GPT-6 sol | 3 | 3 | 2.8 | |
| Codex / GPT-6 sol | 3 | 3 | 3.6 | |
| Proto / DeepSeek V4.1 Flash | 3 | 3 | 2.7 |
A separate campaign under the same frozen scorer, not pooled with the release. Conditions and data.