Skip to content

opensite-ai/opensite-skills / book-recorded-sessions

book-recorded-sessions

Verified

Plan, record, and merge recorded sessions into book chapters: solo lessons, interviews with a proxy reader, multi-guest panels, or hybrid lesson-plus-conversation episodes that also serve video and short-form clips. Covers contract-driven run-of-show, contributor releases and employer clearance, per-speaker capture and transcripts, speaker-attributed atoms, merging many sessions into a traceable chapter, and clip hand-off flags. Use when chapters are built from spoken recordings or podcasts. Not for narrating the finished audiobook or editing clips.

Recorded sessions to chapters

A good conversation is not a chapter. Design each recording from the chapter contract, capture every speaker on their own track, and merge by tracing every paragraph back to someone's recorded words. The author teaches; guests contribute cases, and their words are credited.

Procedure

  1. Pick the format per chapter using references/session-formats.md: solo, interview (author plus a proxy reader), panel, or hybrid (part A lesson with a proxy reader, part B conversation with practitioners). Default for a teaching book with guests: hybrid.
  2. Contributors and rights first. List everyone in 01-evidence/CONTRIBUTORS.csv (role, attribution, speaker labels, rights_id). Before recording, each guest signs a release and the employer clears the session when recorded on company time or premises or about company work: references/contributor-release.md. Track each release as a RIGHTS_LEDGER.csv row.
  3. Plan. Register the session in 01-evidence/SESSIONS.csv, then python scripts/session_plan.py <root> --chapter ch01 --format hybrid --session-id S01 --out 01-evidence/raw/sessions/S01/RUN_OF_SHOW.md. It fails if the chapter contract is incomplete or a needed role is missing.
  4. Record to the capture spec in references/session-formats.md: one microphone per speaker on its own track, slate, room tone, and segment markers dropped by the producer.
  5. Ingest. Put one transcript per track (<label>.vtt or .srt) and markers.csv (segment_id,start) in the session folder, then python scripts/session_ingest.py <root> --session S01 --watchlist <confidential terms> --json 06-production/reports/session_S01.json. It writes timeline.csv and segments.csv (clip candidates flagged for claim checks and clearance) and reports talk share per speaker.
  6. Atomize into 01-evidence/ATOMS.csv with speaker (contributor_id) and session_id; run book-author-source-capture/scripts/atom_coverage.py. Record short solo pickups for gaps.
  7. Merge chapters with references/merge-rules.md: the author's lesson is the spine; guests' cases appear as credited quotes or credited paraphrase. End every paragraph with <!-- src:A-0012 --> (same paragraph, comma-separated for several atoms). Composed connective text goes in {{ai}}...{{/ai}} for the author to rewrite (mode T).
  8. Trace. python scripts/chapter_trace.py <root> --json 06-production/reports/chapter_trace.json before G4; with --lock --json 06-production/reports/chapter_trace_lock.json before G6.
  9. Clips hand-off. Give segments.csv to the clipping workflow. Do not publish a segment flagged needs_claim_check until its claims pass book-evidence-factcheck, or one flagged needs_clearance until the owner clears it.

Acceptance evidence (external)

  • Signed releases and employer clearance on file before any guest's words are used.
  • atom_coverage.json passes from session atoms; chapter_trace passes (draft at G4, lock at G6).
  • The contributors approve their credited quotes in writing; independent readers judge the chapters in beta rounds. Colleagues on the show are not validators.

Never

  • Put a guest's experience in the author's mouth, or build composite quotes or people.
  • Record or use a guest's words without a signed release; publish client names, deal terms, or internal numbers without written clearance.
  • Treat a transcript cleanup as a chapter: no merge without the contract, trace, and fact-check.

Files

  • scripts/session_plan.py — chapter contract to run-of-show; checks roles, releases, coverage.
  • scripts/session_ingest.py — per-speaker transcripts to timeline, talk share, segment flags.
  • scripts/chapter_trace.py — paragraph-level trace, guest release and credit checks, author share.
  • assets/segment_types.yaml — segment library and format timings used by session_plan.py.
  • references/session-formats.md — formats, roles, capture spec, time budget. Read at step 1.
  • references/merge-rules.md — how to merge sessions into a chapter. Read at step 7.
  • references/contributor-release.md — release and clearance checklist. Read at step 2.
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall