Rapid context extractor
Skill thevibethinker/vibe-thinker-skills/rapid-context-extractor
Portable, standalone Zo Computer skills you can install, reuse, and share.
npx -y skills add thevibethinker/vibe-thinker-skills --skill rapid-context-extractorAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 1 stars1 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
Extract and teach key points from a source using seed context, then force active engagement. Use when analyzing articles, documents, transcripts, video/audio transcripts, or mixed media where output must preserve chronological idea flow, include concept explanation, and prompt user reflection.
SKILL.md
3.6 KB, 729 tokens by cl100k_base, as published. Nobody here has run it
Rapid Context Extractor
Normalize source mechanics with the script, then do semantic analysis in chat.
Quick Start
python3 Skills/rapid-context-extractor/scripts/prepare_payload.py \
--seed-file "./Research/topic-frame.md" \
--source-url "https://example.com/article" \
--auto-semantic \
--output "/home/.z/workspaces/<conversation-id>/extraction_packet.md"
Replace <conversation-id> with your active conversation workspace, or use any other writable output path.
Use one source input per run:
--source-urlfor web pages--source-filefor local docs/transcripts/subtitles--source-textfor pasted text
Optional seed context:
--seed-fileor--seed-text
Optional semantic memory anchoring:
--semantic-queryto retrieve relevant prior concepts from your semantic memory--auto-semanticto generate semantic query from source title + extracted terms (recommended default)--semantic-limit(default 5) to control number of memory anchors--provenanceto force frontmatter provenance (otherwise inferred from output path conversation ID)
Workflow
- Prepare packet
- Run
scripts/prepare_payload.pyto produce a markdown packet containing seed context + chronological source chunks. - For media files, require transcript sidecar or transcribe first.
- Adopt analyst frame
- Read seed context first.
- State the frame in 1-2 lines before distillation.
- If missing background blocks understanding, perform targeted research before summarizing.
- Distill in chronological order
- Produce bullet points in the order ideas appear in source.
- Avoid regrouping by theme if it breaks chronology.
- Keep claims faithful to the source.
- Include image meaning
- If visuals exist, summarize what each visual contributes to the argument.
- Note if visuals reinforce, contradict, or extend text claims.
- Integrate for learning
- Explain key terms, concepts, and implications in plain language.
- Connect key claims to
Semantic Memory Anchorswhere relevant (agreements, tensions, extensions). - Explicitly classify each integration claim as
aligns,extends, orconflicts/tension. - Ask clarifying questions that advance interpretation or decisions.
- Force active engagement
- Ask for immediate reaction (1-3 lines acceptable).
- Ask for one agreement and one challenge.
- Offer optional ingestion: only ingest if user explicitly says yes.
Standard Output Shape
Use this structure in responses:
Analytical FrameChronological DistillationVisual Layer(if applicable)Semantic Integration(link to user- or project-specific anchors when available) : include explicitaligns/extends/conflictslabelsConcept DecoderClarifying QuestionsYour Reaction(collect user response)Optional Next Step(ingest yes/no)
Content Library Ingestion
Only after explicit approval, and only in workspaces that include the N5 ingestion helper:
python3 N5/scripts/content_ingest.py "<artifact_path>" --move
Confirm with: Ingested to Content Library as <type>.
Resources
scripts/prepare_payload.py: deterministic intake/normalization for seed + source.references/output-template.md: copyable response template for consistent execution.
What ships with it: 3 files
19.7 KB alongside SKILL.md, 1 of them executable
agents/
- openai.yaml237 B
references/
- output-template.md803 B
scripts/
- prepare_payload.pyruns18.7 KB