Skip to main content Accueil Créateurs praxstack skills-and-personas transcript-pipeline
transcript-pipeline This skill should be used when the user asks to "process this transcript", "convert lecture to notes", "run transcript pipeline", "generate class tutorial from Zoom captions", "validate transcript coverage", or "enrich class resources" (Notion/Canva/Drive links) for bootcamp notes.
Aller à l'installation Skills Marketplace Découvrez et explorez les compétences IA créées par la communauté.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Afficher les détails du prompt Une commande directe contourne le prompt de vérification. Examinez la source avant de l'exécuter.
npx skills add https://github.com/praxstack/skills-and-personas --skill transcript-pipelineLa commande reste sur une seule ligne. Faites défiler horizontalement pour la vérifier avant de la copier.
Vous préférez une copie locale ? Téléchargez les fichiers actuellement disponibles dans SkillsMP.
Télécharger Zip Téléchargement...
Create exhaustive BLUEPRINT.md files — the implementation bible that expands a SPEC.md into maximum granular detail so an AI coding agent can translate it line-by-line into code with zero interpretation. This skill REQUIRES a SPEC.md to already exist. Use this skill whenever the user wants to create a blueprint, implementation blueprint, detailed spec, "bible" document, or says "create the blueprint", "expand the spec", "make it more detailed", "generate BLUEPRINT.md", or "I need every detail spelled out." Also trigger when the user has a SPEC.md and wants to go deeper before implementation. The Blueprint is the final artifact before code — it takes the portable implementation contract (SPEC.md) and fills in every edge case, validation rule, error message, sequence, and example so nothing is left to judgment. Works for any software domain. If no SPEC.md exists, this skill MUST redirect the user to generate one first using the spec-creator skill.
Create production-grade SPEC.md files — portable implementation contracts that any AI coding agent can build from without follow-up questions. Use this skill whenever the user wants to write a spec, specification, technical spec, implementation contract, SPEC.md, or service specification for any software task. Also trigger when the user says "spec out", "write a spec for", "create a spec", "define the contract for", or describes a system they want built and needs a thorough specification before implementation. This skill works for backend services, full-stack apps, CLI tools, SDKs, developer tools, and any domain. It produces specs at the level of a principal-engineer-authored portable implementation contract — not a PRD, not an HLD, but the artifact that sits between architecture and code.
Scope-calibrated ultra-rigor workflow for hard reasoning, high-risk code changes, architecture decisions, debugging with multiple plausible root causes, security-sensitive work, and user requests like "ultra reasoning", "think harder", "verify everything", "adversarial review", "war room", "deep check", "paranoid verifier", or "no hallucinations". Use to force evidence-first planning, assumption tracking, hypothesis falsification, adversarial self-review, verification gates, and clear uncertainty without overloading trivial tasks.
Explorateur de fichiers
38 fichiers name transcript-pipeline description This skill should be used when the user asks to "process this transcript", "convert lecture to notes", "run transcript pipeline", "generate class tutorial from Zoom captions", "validate transcript coverage", or "enrich class resources" (Notion/Canva/Drive links) for bootcamp notes. triggers ["run transcript pipeline","generate class tutorial","validate transcript coverage","enrich class resources"] version 0.2.0
Transcript Pipeline Skill
Run a deterministic, auditable transcript-to-tutorial workflow with optional resource enrichment.
Purpose
Use this skill to convert raw class captions into high-quality study notes while preserving accountability through ledger + validation artifacts.
Use scripts for deterministic work. Use chat/stage prompts for language-heavy transformation.
Core Contract
Keep stage order: ingest -> refine -> synthesize -> enhance -> validate -> publish.
Run deterministic gates with scripts, never with LLM self-certification.
Preserve traceability in .pipeline/* artifacts.
Keep learner-facing notes readable and sanitized.
Treat validation status as PASS/FAIL source of truth.
Scripts
Use these scripts from scripts/:
ingest_zoom_captions.py - deterministic ingestion and segment ledger creationrun_chat_pipeline.py - guided orchestration for stage handoffs and validation
validate_coverage.py - hard-gate coverage validation
publish_tutorial_notes.py - learner-facing file naming and sanitization
merge_chunks.py - merge chunk outputs for large transcripts
run_colab_notebook_pipeline.py - AI/ML Colab appendix and code explainer pipeline
update_ai_notes_with_resources_and_colab.py - AI/ML notes enrichment utility
resource_enrichment.py - authenticated enrichment for Notion/Canva/Drive resources
Stage Workflow
Stage 0: Ingest (Deterministic) python scripts/ingest_zoom_captions.py "<transcript_or_session_path>"
.pipeline/segment_ledger.jsonl.pipeline/segment_manifest.jsonl
Stage 1: Refine (Chat Stage) Load references/stage1-refine.md.
.pipeline/refined_transcript.md.pipeline/topic_inventory.json.pipeline/corrections_log.csv.pipeline/uncertainty_report.json
Stage 2: Synthesize (Chat Stage) Load references/stage2-synthesize.md.
.pipeline/structured_notes.md.pipeline/coverage_matrix.json
Stage 3: Enhance (Chat Stage)
references/stage3-enhance.mdreferences/tutorial-tech-bar-raiser.md
.pipeline/enhanced_notes.mdfinal_notes.mdbootcamp_index.md
Stage 4: Validate (Deterministic) python scripts/validate_coverage.py --pipeline-dir .pipeline
Validation guidance: references/stage4-validate.md.
Segment coverage accountability
Uncertainty retention
No orphan claims
Stage 5: Publish python scripts/publish_tutorial_notes.py --root "<sessions_root>" --session-dir "<session_dir>"
Published tutorial filename in canonical format
Learner-safe note without noisy source tags
Updated course index links
One-Command Guided Mode Use guided runner for chat-window workflows:
python scripts/run_chat_pipeline.py run "<transcript_or_session_path>" --deep-pass
This enforces required handoffs and deep quality gates.
Optional Resource Enrichment Stage Run when class notes include external links (Notion/Canva/Drive):
python scripts/resource_enrichment.py --all-sessions
python scripts/resource_enrichment.py --session-dir "<session_dir>"
Notion: NOTION_TOKEN_V2, NOTION_ACTIVE_USER
Canva: RESOURCE_PLAYWRIGHT_STORAGE_STATE
Reference: references/resource-enrichment-authenticated-flow.md.
Optional AI/ML Colab Enrichment Run for Colab-backed AI/ML classes:
python scripts/run_colab_notebook_pipeline.py
Reference: references/colab-notebook-explainer-pipeline.md.
Large Transcript Handling If input exceeds context comfort:
Run Stage 1 by chunks.
Merge chunk artifacts:
python scripts/merge_chunks.py --chunk-dirs "<chunkA/.pipeline>" "<chunkB/.pipeline>" --output-dir "<session/.pipeline>"
Continue Stage 2 onward on merged artifacts.
Required Outputs Checklist
final_notes.md<Domain> Class <NN> [DD-MM-YYYY] - <Topic>.mdbootcamp_index.md
.pipeline/segment_ledger.jsonl.pipeline/segment_manifest.jsonl.pipeline/refined_transcript.md.pipeline/topic_inventory.json.pipeline/corrections_log.csv.pipeline/uncertainty_report.json.pipeline/structured_notes.md.pipeline/coverage_matrix.json.pipeline/enhanced_notes.md.pipeline/validation_report.md.pipeline/exceptions.json (if fail)
.pipeline/deep_pass_report.md (when --deep-pass).pipeline/deep_pass_exceptions.json (when --deep-pass) Resource enrichment (optional):
.resources/resource_enrichment_report.json
Execution Rules
Fail fast on missing required artifacts.
Report missing outputs explicitly by file path.
Retry only from earliest failing stage.
Keep resource extraction status explicit (success/fallback/blocked).
