Skip to content
Published studies are shown for demonstration purposes, not because they are good or bad research. Demonstrator · items marked Example are invented · what exists today
Search

How has the monthly mean air temperature at 2 metres in Frankfurt am Main (NA...

How this study was produced: every step of the run with the file it wrote, and everything the run used, pinned to the exact version. The address of this page is derived from its content: if anything below changed, the address would change.

DossierDemonstrationVersion 1 of the studyOpen the study
Dossiersha256:a6ba92eccb530a5b395e50b8e3797a41859dbb1ce55b787f5844eff431505409
Dossier as JSON

Paper, data and code of the study

  • Paper
    Not public.

  • Data
    Not public.

  • Code
    Not public.

Run settings

e2er
version 0.16.0 · code at
Template
time-series-forecasting · iterative
Governance
full
AI backend
claude_code
Models
claude_code/sonnet
Exported
20261015
Sessions
6 sessions · listed with the workflow
Outcome
Completed · 2026-10-13 22:20 UTCe2er's internal quality review: 5.9 of 10
e2er code
no local changes at run time

Input data

SHA-256 of the data files, as recorded in the study's provenance.

4 data files with their hash values
  • ·data/data.db 36eec22a2b7c
  • ·data/data_dictionary.json 8ebfa4eee548
  • ·data/data_sources.json 13647c22e4e6
  • ·data/data_summary.md 22beac907d9c

Data used

NASA Prediction Of Worldwide Energy Resources (POWER), NASA Langley Research Center · 1 load · release v2.10.0 · read 2026-10-13

e2er's connector for it · about this data source

NASA POWER monthly T2M at lat 50.11, lon 8.68, 1981–2025 · release v2.10.0, read 2026-10-13 · table frankfurt_t2m_monthly https://power.larc.nasa.gov/api/temporal/monthly/point?parameters=T2M&community=RE&latitude=50.11&longitude=8.68&start=1981&end=2025&format=JSON 3471530b3a04

Terms Free and keyless; NASA data without a marked restriction are CC0. POWER asks publications to give its reference with the API version and access date, and asks to be told when data are passed to others. Full terms

Cite NASA Langley Research Center (LaRC) POWER Project. Prediction Of Worldwide Energy Resources (POWER), https://power.larc.nasa.gov/. The data was obtained from National Aeronautics and Space Administration (NASA) Langley Research Center's Prediction Of Worldwide Energy Resources (POWER) project funded through the NASA Earth Science Division. Data reference: The data was obtained from the POWER Monthly and Annual API version 2.10.0 on 2026/10/13.

References (12)

The works the paper cites, as the dossier records them: 0 from the authors' papers, 1 found on the web by the literature search, 11 method sources from the template.

A method source is a work the template credits for a method it uses. e2er adds it to the study's bibliography; the paper cites the ones its study used.

12 references· 0 from the authors' papers, 1 found on the web, 11 method sources from the template
  1. Prediction Of Worldwide Energy Resources (POWER) found on the web
  2. Nonparametric Tests Against Trend (1945) method source · doi:10.2307/1907187
  3. Estimates of the Regression Coefficient Based on Kendall's Tau (1968) method source · doi:10.1080/01621459.1968.10480934
  4. Distribution of the Estimators for Autoregressive Time Series with a Unit Root (1979) method source · doi:10.1080/01621459.1979.10482531
  5. Testing the null hypothesis of stationarity against the alternative of a unit root (1992) method source · doi:10.1016/0304-4076(92)90104-Y
  6. Forecasting: Principles and Practice (3rd edition) (2021) method source
  7. Time Series Analysis: Forecasting and Control (4th edition) (2008) method source · doi:10.1002/9781118619193
  8. Automatic Time Series Forecasting: The forecast Package for R (2008) method source · doi:10.18637/jss.v027.i03
  9. Out-of-sample tests of forecasting accuracy: an analysis and review (2000) method source · doi:10.1016/S0169-2070(00)00065-0
  10. Another look at measures of forecast accuracy (2006) method source · doi:10.1016/j.ijforecast.2006.03.001
  11. Comparing Predictive Accuracy (1995) method source · doi:10.1080/07350015.1995.10524599
  12. Testing the equality of prediction mean squared errors (1997) method source · doi:10.1016/S0169-2070(96)00719-4

e2er's internal quality review

Six reviewer specialists each score the draft from one angle, on a scale of 0 to 10. The score in the middle is the weighted average of these scores, with the weights in the table.

data 6.5 literature 4.6 methods 6.0 plausibility 6.5 technical 6.0 writing 5.5 5.9 of 10

5.9 of 10

Each reviewer's score and weight
ReviewerScoreWeight
data6.51
literature4.60.75
methods6.01.5
plausibility6.51
technical6.01.25
writing5.50.75

Pre-registration

The design was written down and frozen before the analysis ran. The hash value is the SHA-256 of the file at that moment.

File
design/preregistration.md
Hash value
06aa1494d793
Frozen
2026-10-13T21:52:59Z
Plan files
  • forecast_design.json 73cdec2dd9c7

The plan files were frozen with the pre-registration; the hash value after each is the SHA-256 of the file at that moment. The pre-registration check compares the files with these hash values.

Rounds of improvement (1)

The run worked on its draft in rounds. After each round a ceiling check decided whether another round would help, whether the study needed a change of approach, or whether it was ready for review. The steps of each round are marked with the round in the workflow below.

  1. Round 1 · Estimation, Paper draft, Abstract

    Data and forecast design are done and frozen. The only open step is the researcher-mandated layout fix to estimation_results.json (h1_result, h2_result, h3_result as top-level dicts). After that, the evaluation check runs and the draft and abstract can follow.

    Decision after round 1ready for reviewThe draft is already complete and internally coherent: the forecast setup was frozen before estimation (train 1981-2022, 36-month hold-out 2023-2025), all declared models are reported against the seasonal naive and naive baselines, and the results cover trend, seasonal trends, amplitude and phase, Diebold-Mariano tests and interval calibration. The remaining gaps are the short hold-out, which limits power, and the grid-cell rather than station nature of the data. Targeted work cannot fix either of these, because the hold-out is frozen and the data are fixed. Further drafting on iteration 1 would give less than 5% qualitative improvement, and the review panel (technical, literature, writing, data, methods, plausibility) is better placed to find residual issues, such as checking the reported ~30 percent RMSE reduction against estimation_results.json and the rolling-origin evidence behind the claims.

Workflow (56 steps)

The run in order, from the run's own record. The run was carried out in 6 sessions. How to read it

A session is one e2er process, from a start or a resume to the next stop, and each session names the e2er version and commit its steps ran on. Each specialist step shows the model, the file it wrote and the file's hash value (SHA-256). Where the run did not record the hash value, the step says so: the last step that wrote a file shows the hash value of the file at export, and earlier steps that wrote the same file are marked as superseded. Checks that stopped a step are listed where they happened: 7 here. Steps taken by the researcher (7 here) are marked with a grey dot. Stops, pauses and failures of the run are marked with a ring, at their time.

Session 1 of 6 2026-10-13 21:27–21:32 UTC · e2er 0.16.0 · 7 steps · literature scanner, idea developer, data architect and 2 more · 2 researcher actions · Run stopped for the researcher's review (review design)
  1. Run: backend identity · 21:27
  2. Run: method sources · 21:27
  3. Initial
  4. literature scanner · sonnet · 21:27–21:28 design/literature_review.md cc9db2990c62
  5. idea developer · sonnet · 21:27–21:28 design/paper_plan.md 12edb685802f
  6. data architect · sonnet · 21:28–21:29 data/data_dictionary.json eb1588b8e8d5
  7. data analyst · sonnet · 21:29–21:30 data/data_summary.md 22beac907d9cresults/summary_statistics.json 0104f829e880results/figure_spec.json bbe5bb2aa684
  8. forecast designer · sonnet · 21:30–21:31 design/forecast_design.md 27558e264741design/forecast_design.json 73cdec2dd9c7
  9. Run stopped for the researcher's review (review design) · 21:31
  10. Researcher gave an instruction for review design · 21:32 Researcher decisions at the design review: the forecast setup in forecast_design.json (training 1981-01 to 2022-12, hold-out 2023-01 to 2025-12, seasonal naive and naive baselines, ETS(A,N,A), ETS(A,Ad,A), SARIMA, harmonic trend) is the design; the forecast beyond the data covers 2026-01 to 2028-12. Drop the optional ML model and the DWD station comparison (no DWD data are loaded in this study); name the station comparison as future work only. The data end in 2025-12 because the POWER monthly load covers whole years; say so. Report every declared model, including those that lose to the seasonal naive baseline.
  11. Researcher approved review design · 21:32
Session 2 of 6 2026-10-13 21:32–21:52 UTC · e2er 0.16.0 · 3 steps · 2 researcher actions · Run stopped for the researcher's review (preregister)
  1. Initial
  2. Check: preregistration passed
  3. Run: backend identity · 21:32
  4. Run stopped for the researcher's review (preregister) · 21:32
  5. Researcher edited preregistration.md · 21:52 a05acdae61bf → 06aa1494d793
  6. Researcher approved preregister · 21:52
Session 3 of 6 2026-10-13 21:52–21:59 UTC · e2er 0.16.0 · 10 steps · econometrics specialist · 1 researcher action · 4 checks failed · Pre-registration frozen
  1. Initial
  2. Check: preregistration passed
  3. Check: forecast design passed
  4. Run: backend identity · 21:52
  5. Pre-registration frozen · 21:52 hash value 06aa1494d793
  6. Check: contracts failed, step stopped
    What the check foundestimation_results.json: pre-registration: pre-registered hypotheses without a result: H1, H2, H3 (each result entry names the hypothesis it tests in a 'hypothesis' field). The hypotheses and sample size are frozen in preregistration.lock.json (and in the machine-readable block of preregistration.md).
  7. econometrics specialist · sonnet · 21:52–21:55 design/econometric_spec.md 08430c68dcedresults/estimation_results.json ccb4dfdfdd5e
  8. Check: contracts failed, step stopped
    What the check foundestimation_results.json: pre-registration: pre-registered hypotheses without a result: H1, H2, H3 (each result entry names the hypothesis it tests in a 'hypothesis' field). The hypotheses and sample size are frozen in preregistration.lock.json (and in the machine-readable block of preregistration.md).
  9. econometrics specialist · sonnet · 21:55–21:56 design/econometric_spec.md a5615d45a665results/estimation_results.json e4c206661a26
  10. Check: contracts failed, step stopped
    What the check foundestimation_results.json: pre-registration: pre-registered hypotheses without a result: H1, H2, H3 (each result entry names the hypothesis it tests in a 'hypothesis' field). The hypotheses and sample size are frozen in preregistration.lock.json (and in the machine-readable block of preregistration.md).
  11. econometrics specialist · sonnet · 21:56–21:58 design/econometric_spec.md a5615d45a665 (unchanged by this step)results/estimation_results.json 25a6ace9e350
  12. Check: output contract failed, run stopped
    What the check foundeconometrics specialist: the output failed its check in 3 attemptsAttempt 1: estimation_results.json: pre-registration: pre-registered hypotheses without a result: H1, H2, H3 (each result entry names the hypothesis it tests in a 'hypothesis' field). The hypotheses and sample size are frozen in preregistration.lock.json (and in the machine-readable block of preregistration.md).Attempt 2: estimation_results.json: pre-registration: pre-registered hypotheses without a result: H1, H2, H3 (each result entry names the hypothesis it tests in a 'hypothesis' field). The hypotheses and sample size are frozen in preregistration.lock.json (and in the machine-readable block of preregistration.md).Attempt 3: estimation_results.json: pre-registration: pre-registered hypotheses without a result: H1, H2, H3 (each result entry names the hypothesis it tests in a 'hypothesis' field). The hypotheses and sample size are frozen in preregistration.lock.json (and in the machine-readable block of preregistration.md).
  13. Researcher sent back econometrics specialist · 21:59 Your results do name H1, H2 and H3, but e2er 0.16.0 reads the 'hypothesis' field only on dictionaries at the top level of estimation_results.json; a top-level list (hypothesis_results) and the nested trend entries are not read (an e2er defect, recorded by the researcher). Change only the layout, no number: replace the list hypothesis_results by three top-level dictionaries h1_result, h2_result and h3_result, each holding the content of its former list entry with its 'hypothesis' field, and update any table_spec or figure paths that pointed into hypothesis_results. Keep everything else as it is.
Session 4 of 6 2026-10-13 21:59–22:08 UTC · e2er 0.16.0 · 13 steps · econometrics specialist, abstract writer, paper drafter · 1 researcher action · 3 checks failed · Run: backend identity
  1. Run: backend identity · 21:59
Initial 2 steps · econometrics specialist · 21:59–22:00
  1. econometrics specialist · sonnet · 21:59–22:00 Run again after the researcher sent it back design/econometric_spec.md 728034061cfbresults/estimation_results.json d98ac474f5e6
  2. Check: forecast evaluation passed
Iterative 5 steps · econometrics specialist, abstract writer, paper drafter · 22:00–22:03 · 1 check failed
  1. econometrics specialist · sonnet · 22:00–22:01 Round 1 design/econometric_spec.md 728034061cfb (unchanged by this step)results/estimation_results.json d98ac474f5e6 (unchanged by this step)
  2. abstract writer · sonnet · 22:01–22:01 Round 1 paper/abstract.tex f0bef6b8b8f9
  3. Check: contracts failed, step stopped
    What the check foundpaper_draft.tex: 1 construction(s) the researcher does not accept in a paper: "Against this background" in: Against this background, this paper addresses two descriptive and predictive questions. (pivot to the contribution directly (for instance: To keep these quantities apart, we propose ...)). Rewrite each of these sentences without the construction; keep its content
  4. paper drafter · sonnet · 22:01–22:03 Round 1 paper/paper.tex 689c5e2fc695results/table_spec.json c0706427a4c3
  5. paper drafter · sonnet · 22:03–22:03 Round 1 paper/paper.tex e8a447f6e886results/table_spec.json c0706427a4c3 (unchanged by this step)
Estimation gate 2 steps · 22:03
  1. Check: preregistration passed No change since the pre-registration was frozen (2026-10-13 21:52 UTC).
  2. Check: estimation passed
Forecast gate 1 step · 22:03
  1. Check: forecast evaluation passed
Figure gate 3 steps · 22:03–22:08 · 1 researcher action · 2 checks failed
  1. Check: figure data failed, step stopped fig_seasonal_profile.pdf names no 'source' (a data.db table and columns, or a results path)
  2. Check: figure gate failed, run stopped fig_seasonal_profile.pdf names no 'source' (a data.db table and columns, or a results path)
  3. Researcher sent back data analyst · 22:08 Only one fix: in figure_spec.json, fig_seasonal_profile.pdf has no 'source'. Give it the source its bars come from (the 1981-2010 calendar-month climatology: a data.db table with its columns, or the path in summary_statistics.json), at the precision the values are stated. Do not change the data, the other figures or any result.
Session 5 of 6 2026-10-13 22:08–22:09 UTC · e2er 0.16.0 · 7 steps · data analyst · 1 researcher action · Run stopped for the researcher's review (review draft)
  1. Figure gate
  2. Check: forecast evaluation passed
  3. Run: backend identity · 22:08
  4. data analyst · sonnet · 22:08–22:08 Run again after the researcher sent it back data/data_summary.md 22beac907d9c (unchanged by this step)results/summary_statistics.json 0104f829e880 (unchanged by this step)results/figure_spec.json 96226e2de204
  5. Estimation gate
  6. Check: preregistration passed No change since the pre-registration was frozen (2026-10-13 21:52 UTC).
  7. Check: estimation passed
  8. Forecast gate
  9. Check: forecast evaluation passed
  10. Figure gate
  11. Check: figure data passed
  12. Run stopped for the researcher's review (review draft) · 22:08
  13. Researcher approved review draft · 22:09
Session 6 of 6 2026-10-13 22:09–22:20 UTC · e2er 0.16.0 · 16 steps · polish formula, patch revisor, literature reviewer and 6 more · Polish notes considered
Estimation gate 2 steps · 22:09
  1. Check: preregistration passed No change since the pre-registration was frozen (2026-10-13 21:52 UTC).
  2. Check: estimation passed
Forecast gate 1 step · 22:09
  1. Check: forecast evaluation passed
Figure gate 1 step · 22:09–22:10 · Self-critique
  1. Check: figure data passed
  2. Run: backend identity · 22:09
  3. Self-critique · 22:10 0 findings, 0 serious
Polish 2 steps · polish formula, patch revisor · 22:10–22:11
  1. polish formula · sonnet · 22:10–22:11 reviews/polish_formula.md 301ffc50aed7
  2. patch revisor · sonnet · 22:11–22:11 misc/paper_draft.tex.edits.json 35c571b35028
Review 8 steps · literature reviewer, writing reviewer, data reviewer and 3 more · 22:11–22:16 · Polish notes considered
  1. Check: numbers passed
  2. Polish notes considered · 22:11 1 note; 2 changes made in the draft
  3. Check: citations passed
  4. literature reviewer · sonnet · 22:12–22:12 reviews/review_literature.md c6a11b51acad
  5. writing reviewer · sonnet · 22:12–22:13 reviews/review_writing.md 061e91db30b4
  6. data reviewer · sonnet · 22:12–22:14 reviews/review_data.md 3128ba87481b
  7. technical reviewer · sonnet · 22:12–22:14 reviews/review_technical.md 531b5a528419
  8. methods reviewer · sonnet · 22:13–22:15 reviews/review_methods.md ba27a0d0a735
  9. plausibility reviewer · sonnet · 22:14–22:16 reviews/review_plausibility.md f1e085593a2f
Revision 1 step · patch revisor · 22:16–22:17
  1. patch revisor · sonnet · 22:16–22:17 misc/paper_draft.tex.edits.json a5190b0d942b
Replication 1 step · replication packager · 22:17–22:20
  1. replication packager · sonnet · 22:17–22:20 replication/estimation.py eb6acc1decd2

The run ended: completed · 2026-10-13 22:20 UTC

What it was built with (65)

Each part with its authors and the version used. e2er parts are pinned by git's content hash of their file at the commit above; follow a link to read that exact version.

Templates (1)

· by e2er authors
TemplateByPinned version
time-series-forecastinge2er authors unresolved: the run recorded no E2ER commit

Specialists (17)

· by e2er authors
SpecialistByPinned version
abstract_writere2er authors unresolved: the run recorded no E2ER commit
data_analyste2er authors unresolved: the run recorded no E2ER commit
data_architecte2er authors unresolved: the run recorded no E2ER commit
data_reviewere2er authors unresolved: the run recorded no E2ER commit
econometrics_specialiste2er authors unresolved: the run recorded no E2ER commit
forecast_designere2er authors unresolved: the run recorded no E2ER commit
idea_developere2er authors unresolved: the run recorded no E2ER commit
literature_reviewere2er authors unresolved: the run recorded no E2ER commit
literature_scannere2er authors unresolved: the run recorded no E2ER commit
methods_reviewere2er authors unresolved: the run recorded no E2ER commit
paper_draftere2er authors unresolved: the run recorded no E2ER commit
patch_revisore2er authors unresolved: the run recorded no E2ER commit
plausibility_reviewere2er authors unresolved: the run recorded no E2ER commit
polish_formulae2er authors unresolved: the run recorded no E2ER commit
replication_packagere2er authors unresolved: the run recorded no E2ER commit
technical_reviewere2er authors unresolved: the run recorded no E2ER commit
writing_reviewere2er authors unresolved: the run recorded no E2ER commit

Skills (46)

· by e2er authors
SkillByPinned version
researchere2er authors unresolved: the run recorded no E2ER commit
cleaninge2er authors unresolved: the run recorded no E2ER commit
data-tablese2er authors unresolved: the run recorded no E2ER commit
eurostate2er authors unresolved: the run recorded no E2ER commit
exoplanetse2er authors unresolved: the run recorded no E2ER commit
figure-spece2er authors unresolved: the run recorded no E2ER commit
frede2er authors unresolved: the run recorded no E2ER commit
gaiae2er authors unresolved: the run recorded no E2ER commit
giscoe2er authors unresolved: the run recorded no E2ER commit
githube2er authors unresolved: the run recorded no E2ER commit
gmde2er authors unresolved: the run recorded no E2ER commit
gutenberge2er authors unresolved: the run recorded no E2ER commit
nasa_powere2er authors unresolved: the run recorded no E2ER commit
naturalearthe2er authors unresolved: the run recorded no E2ER commit
noaae2er authors unresolved: the run recorded no E2ER commit
owide2er authors unresolved: the run recorded no E2ER commit
query-datae2er authors unresolved: the run recorded no E2ER commit
summary-statistics-schemae2er authors unresolved: the run recorded no E2ER commit
table-spece2er authors unresolved: the run recorded no E2ER commit
timeseries-results-schemae2er authors unresolved: the run recorded no E2ER commit
usgse2er authors unresolved: the run recorded no E2ER commit
who_ghoe2er authors unresolved: the run recorded no E2ER commit
worldbanke2er authors unresolved: the run recorded no E2ER commit
yfinancee2er authors unresolved: the run recorded no E2ER commit
panel-datae2er authors unresolved: the run recorded no E2ER commit
time-seriese2er authors unresolved: the run recorded no E2ER commit
econ-modele2er authors unresolved: the run recorded no E2ER commit
optimization-verificatione2er authors unresolved: the run recorded no E2ER commit
time-series-forecastinge2er authors unresolved: the run recorded no E2ER commit
anti-slope2er authors unresolved: the run recorded no E2ER commit
creative-ideatione2er authors unresolved: the run recorded no E2ER commit
noveltye2er authors unresolved: the run recorded no E2ER commit
consistency-checke2er authors unresolved: the run recorded no E2ER commit
data-qualitye2er authors unresolved: the run recorded no E2ER commit
domain-plausibilitye2er authors unresolved: the run recorded no E2ER commit
methods-reviewe2er authors unresolved: the run recorded no E2ER commit
referee-simulatione2er authors unresolved: the run recorded no E2ER commit
technical-reviewe2er authors unresolved: the run recorded no E2ER commit
writing-qualitye2er authors unresolved: the run recorded no E2ER commit
context-buildere2er authors unresolved: the run recorded no E2ER commit
replication-packagee2er authors unresolved: the run recorded no E2ER commit
abstracte2er authors unresolved: the run recorded no E2ER commit
cite-numbers-by-sourcee2er authors unresolved: the run recorded no E2ER commit
paper-structuree2er authors unresolved: the run recorded no E2ER commit
personal-stylee2er authors unresolved: the run recorded no E2ER commit
scoped-revisione2er authors unresolved: the run recorded no E2ER commit

Connectors (1)

· by e2er authors
ConnectorByPinned version
claude-codee2er authors unresolved: the run recorded no E2ER commit

Sources of the template “time-series-forecasting”

The work the template of this run is based on, draws from or cites, as the template file names it. A source says where an idea came from; it does not say its authors endorse this study.

Methods

George E. P. Box, Gwilym M. Jenkins and Gregory C. Reinsel. Time Series Analysis: Forecasting and Control (4th edition). Wiley.

Rob J. Hyndman and George Athanasopoulos. Forecasting: Principles and Practice (3rd edition). OTexts, Melbourne.

Rob J. Hyndman and Yeasmin Khandakar. Automatic Time Series Forecasting: The forecast Package for R. Journal of Statistical Software 27(3).

Rob J. Hyndman and Anne B. Koehler. Another look at measures of forecast accuracy. International Journal of Forecasting 22(4), 679-688.

Leonard J. Tashman. Out-of-sample tests of forecasting accuracy: an analysis and review. International Journal of Forecasting 16(4), 437-450.

Francis X. Diebold and Roberto S. Mariano. Comparing Predictive Accuracy. Journal of Business & Economic Statistics 13(3), 253-263.

David Harvey, Stephen Leybourne and Paul Newbold. Testing the equality of prediction mean squared errors. International Journal of Forecasting 13(2), 281-291.

David A. Dickey and Wayne A. Fuller. Distribution of the Estimators for Autoregressive Time Series with a Unit Root. Journal of the American Statistical Association 74(366a), 427-431.

Denis Kwiatkowski, Peter C.B. Phillips, Peter Schmidt and Yongcheol Shin. Testing the null hypothesis of stationarity against the alternative of a unit root. Journal of Econometrics 54(1-3), 159-178.

Henry B. Mann. Nonparametric Tests Against Trend. Econometrica 13(3), 245-259.

Pranab Kumar Sen. Estimates of the Regression Coefficient Based on Kendall's Tau. Journal of the American Statistical Association 63(324), 1379-1389.

Cite e2er with this dossier

A paper produced with e2er gives one reference for e2er: this dossier. It credits everyone listed above.

@misc{e2er-a6ba92eccb530a5b,
  title        = {e2er dossier for "How has the monthly mean air temperature at 2 metres in Frankfurt am Main (NA..."},
  author       = {{e2er}},
  year         = {2026},
  howpublished = {\url{https://e2er.org/d/a6ba92eccb530a5b}},
  note         = {Workflow, templates, specialists, skills, connectors and models used, pinned by content hash. Dossier sha256:a6ba92eccb530a5b395e50b8e3797a41859dbb1ce55b787f5844eff431505409}
}