Calculate change in rank for PyPI dependencies since 2020 to 2026, and from 2024 to 2026
Uses rankings of direct dependencies for all PyPI packages that were first released in each year range
which are pre-computed in the gists (see get_gists.sh to download)
| package,github_repo,uses_trusted_publishing,has_pypi_publish_action,has_id_token_write,tp_signals,publishing_workflow,pinning_status,sha_pinned_actions,total_actions,workflow_count,source,error | |
| certifi,certifi/python-certifi,True,True,True,"id-token:write, pypa/gh-action-pypi-publish, pypi-publish-no-password",release.yml,ALL_SHA,5,5,1,cache, | |
| typing-extensions,python/typing_extensions,True,True,True,"id-token:write, pypa/gh-action-pypi-publish, pypi-publish-no-password",publish.yml,MIXED,11,37,3,cache, | |
| idna,kjd/idna,True,True,True,"id-token:write, pypa/gh-action-pypi-publish, pypi-publish-no-password",deploy.yml,ALL_SHA,9,9,1,cache, | |
| charset-normalizer,jawah/charset_normalizer,True,True,True,"id-token:write, pypa/gh-action-pypi-publish, pypi-publish-no-password",cd.yml,MIXED,38,43,4,cache, | |
| pip,pypa/pip,True,True,True,"id-token:write, pypa/gh-action-pypi-publish, pypi-publish-no-password",release.yml,MIXED,4,21,4,cache, | |
| click,pallets/click,True,True,True,"id-token:write, pypa/gh-action-pypi-publish, pypi-publis |
| rank | package | repo | pinning | tp | publishing_wf | |
|---|---|---|---|---|---|---|
| 1 | boto3 | boto/boto3 | MIXED | False | ||
| 2 | packaging | pypa/packaging | ALL_SHA | False | ||
| 3 | urllib3 | urllib3/urllib3 | ALL_SHA | False | ||
| 4 | setuptools | pypa/setuptools | NONE | False | ||
| 6 | requests | psf/requests | ALL_SHA | False | ||
| 8 | botocore | boto/botocore | MIXED | False | ||
| 11 | aiobotocore | aio-libs/aiobotocore | ALL_SHA | False | ||
| 12 | python-dateutil | dateutil/dateutil | NONE | False | publish.yml | |
| 13 | six | benjaminp/six | NONE | False |
You are directive. When I give you a list of tasks or describe what I need to do, your job is to triage by time sensitivity and then tell me what to do in what order. Give commands, not suggestions. Say "do X" not "you could do X" or "you might want to do X."
You are calm. Do not use urgency language like "right now", "immediately", "ASAP", or "hurry." I don't have a procrastination problem — I have a sequencing and prioritisation problem. Your commands should feel like a composed, clear-headed boss delegating, not a drill sergeant.
You do not hedge inside commands. Once you've decided something belongs in the sequence, commit to it. "Do A, then do B" — not "do A, and then maybe B if you feel up to it." Hedging inside a command undermines the whole point of giving one.
You are not curt. Being directive does not mean being short with me. You can explain your reasoning for the triage order, flag subtasks I mentioned but might forget, and generally be warm. Just
| Category | Oral / Literate | Avg F1 (≈) | Individual Markers (F1) | Comment on Why / Causes | Verdict |
|---|---|---|---|---|---|
| Address & Interaction | Oral | 0.604 | vocative (.675), imperative (.606), second_person (.549), inclusive_we (.608), rhetorical_question (.661), phatic_check (.634), phatic_filler (.495) | Strong lexical and syntactic cues; short-range dep |
| import json | |
| import torch | |
| from huggingface_hub import hf_hub_download | |
| from transformers import AutoModel, AutoTokenizer | |
| def main(): | |
| model_name = "HavelockAI/bert-token-classifier" |
| marker | precision | recall | f1-score | support | |
|---|---|---|---|---|---|
| B-literate_list_structure | 0.975 | 0.75 | 0.848 | 52 | |
| O | 0.771 | 0.847 | 0.807 | 37244 | |
| B-oral_imperative | 0.753 | 0.805 | 0.778 | 87 | |
| B-literate_footnote_reference | 0.81 | 0.739 | 0.773 | 23 | |
| B-oral_rhetorical_question | 0.649 | 0.809 | 0.72 | 89 | |
| I-literate_technical_abbreviation | 0.687 | 0.731 | 0.709 | 108 | |
| B-oral_inclusive_we | 0.603 | 0.793 | 0.685 | 266 | |
| I-literate_footnote_reference | 0.57 | 0.821 | 0.673 | 84 | |
| I-oral_rhetorical_question | 0.646 | 0.683 | 0.664 | 840 |