Loom transcript
Build Your Own AI Employee. The complete starter kit for OpenClaw + Hermes Agent. 155 deduplicated skills, drag-and-drop workspace, case studies, install scripts.
npx -y skills add S3YED/appie-kit --skill loom-transcriptAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 6 stars6 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Fetch the FULL transcript from any Loom share link and turn it into clean markdown, ready to ingest into a knowledge base. Use when given a loom.com/share/... URL and asked to get its transcript, notes, or text.
SKILL.md
2.1 KB, as published. Nobody here has run it
Loom transcript fetch
Loom's transcript file sits behind a SIGNED CloudFront URL, so the plain
cdn.loom.com/.../transcription/<id>.json returns "Missing Key-Pair-Id". The
trick: render the share page in a real browser, capture the signed transcription
request from the network log, then fetch the full transcript JSON from it.
Use it
./loom-to-kb.sh "https://www.loom.com/share/<32-hex-id>" [output-dir]
Prints the path to a clean .md (title + Loom's auto-summary + chapters + the
FULL transcript). Default output dir: ~/clawd/knowledge/loom-transcripts.
How it works (if you need to adapt it)
agent-browser network requests --clearthenagent-browser open <url>— the browser fetches the signed transcript on load.agent-browser network requests --filter transcription→ grab the signedhttps://cdn.loom.com/mediametadata/transcription/<id>-N.json?Policy=...&Signature=...URL.curlthat signed URL → JSON withphrases[], each{ts, value}. Join thevaluefields (NOTtext— that key is empty) for the full transcript.- Title from
agent-browser eval "document.title". Summary/chapters fromagent-browser read.
Dependencies
agent-browser(fleet default browser CLI) — required.python3,curl.
Then ingest into a KB
# fleet shared KB (our own knowledge):
curl -s http://127.0.0.1:8765/cognify/ingest -H 'content-type: application/json' \
-d "{\"tenant\":\"fleet\",\"namespace\":\"loom-calls\",\"title\":\"<title>\",\"text\":\"<md>\"}"
# a client bot's isolated memory: cognify --tenant <client> ingest <file.md>
Gotcha: signed URLs are time-limited (~minutes) — fetch immediately after capture. Loom's transcript panel is virtualized, so reading the DOM only gets a partial; always use the signed-URL path for the full transcript.