| name | substack-extraction |
| description | Export authorized Substack posts into Markdown, media URL manifests, optional media, transcripts, followed Promptkit links, and verification reports using the bundled portable CLI. |
| triggers | ["export substack article","scrape authorized substack","archive substack post","substack transcript","follow grab the prompts"] |
Substack Extraction
Use this skill to archive Substack posts the user can lawfully access. Do not bypass paywalls, DRM, CAPTCHA, or access controls. Use only a user-provided authorized browser session/cookie file or content the user owns/has permission to access.
Bundled CLI
The executable code is self-contained in scripts/.
Install dependency if needed:
python3 -m pip install -r scripts/requirements.txt
Run via the portable launcher:
python3 scripts/substack_extract.py --help
If this repo has been installed as a package, this also works:
substack-extract --help
Single post
python3 scripts/substack_extract.py extract \
--url 'https://example.substack.com/p/post-slug' \
--cookies /path/to/cookies.txt \
--output-root /path/to/substack_exports \
--follow 'Grab the Prompts' \
--follow-domain promptkit.example.com
Add media/transcript only when requested:
python3 scripts/substack_extract.py extract \
--url 'https://example.substack.com/p/post-slug' \
--cookies /path/to/cookies.txt \
--output-root /path/to/substack_exports \
--include-media \
--transcribe
Batch
Copy assets/batch_template.json, fill in URLs/cookie path/output root, then run:
python3 scripts/substack_extract.py batch --config batch.json --continue-on-error
Verify
python3 scripts/substack_extract.py verify EXPORT_DIR --require-links
Use stricter checks when relevant:
python3 scripts/substack_extract.py verify EXPORT_DIR \
--require-media \
--require-transcript \
--require-links \
--fail-on-warning
Outputs
Each export writes index.md, index.html, metadata.json, run_manifest.json, verification_report.md, media URL lists, optional media files, optional transcript files, and optional linked companion pages under linked/.
See references/output-structure.md and references/cookie-handling.md for details.