| name | hdr-dynmeta-docs |
| description | Search and fetch docs for HDR dynamic-metadata CLIs: dovi_tool (Dolby Vision RPU authoring — profiles 4, 5, 7, 8.1, 8.4 — github.com/quietvoid/dovi_tool) and hdr10plus_tool (HDR10+ SEI authoring — github.com/quietvoid/hdr10plus_tool). Canonical subcommand lists (info, generate, editor, export, plot, convert, demux, mux, extract-rpu, inject-rpu for dovi_tool; extract, inject, remove, plot, editor for hdr10plus_tool), HEVC Annex-B + Matroska I/O, integration with ffmpeg bitstream filters + mkvmerge + MP4Box. Use when the user asks to look up a dovi_tool or hdr10plus_tool subcommand, check Dolby Vision profile support, verify HDR dynamic-metadata tool usage, or read the canonical GitHub README.
|
| argument-hint | [query] |
HDR DynMeta Docs
Context: $ARGUMENTS
Quick start
- Find a dovi_tool / hdr10plus_tool subcommand or flag: → Step 2 (
search)
- Read the full README section: → Step 3 (
section)
- Grab a whole README / release-notes page: → Step 4 (
fetch)
- Prime cache for offline use: → Step 5 (
index)
When to use
- User asks "which Dolby Vision profiles does dovi_tool support?" (answer: 4, 5, 7, 8.1, 8.4 — 8.2 unknown).
- User asks about hdr10plus_tool subcommands (extract / inject / remove / plot / editor).
- User is about to chain
ffmpeg bsf=hevc_mp4toannexb | dovi_tool extract-rpu | … and needs to verify flag names.
- Need latest release version and Rust MSRV before recommending install method.
- For the execution side (actually running these tools), see
hdr-dovi-tool + hdr-hdr10plus-tool skills.
Step 1 — Page catalog
uv run ${CLAUDE_SKILL_DIR}/scripts/hdrdocs.py list-pages
| Page | URL |
|---|
dovi-readme | github.com/quietvoid/dovi_tool/blob/main/README.md (raw) |
dovi-releases | api.github.com/repos/quietvoid/dovi_tool/releases/latest |
hdr10plus-readme | github.com/quietvoid/hdr10plus_tool/blob/main/README.md (raw) |
hdr10plus-releases | api.github.com/repos/quietvoid/hdr10plus_tool/releases/latest |
Read references/subcommands.md for the verified subcommand + flag catalog.
Step 2 — Search first
uv run ${CLAUDE_SKILL_DIR}/scripts/hdrdocs.py search --query "extract-rpu" --page dovi-readme
uv run ${CLAUDE_SKILL_DIR}/scripts/hdrdocs.py search --query "editor" --page hdr10plus-readme
Use --format json for machine-parseable output.
Step 3 — Read a section
uv run ${CLAUDE_SKILL_DIR}/scripts/hdrdocs.py section --page dovi-readme --id Usage
Step 4 — Fetch a page
uv run ${CLAUDE_SKILL_DIR}/scripts/hdrdocs.py fetch --page dovi-readme
uv run ${CLAUDE_SKILL_DIR}/scripts/hdrdocs.py fetch --page hdr10plus-releases
Step 5 — Prime the cache
uv run ${CLAUDE_SKILL_DIR}/scripts/hdrdocs.py index
Cache dir: ~/.cache/hdr-dynmeta-docs/. Override with HDR_DYNMETA_DOCS_CACHE=/path.
Gotchas
- Both tools are by
quietvoid, written in Rust. Install via cargo install (from source) or pre-built binaries from each repo's Releases. hdr10plus_tool 1.7.2 (Dec 2025) requires Rust ≥ 1.85.0. dovi_tool tracks similar recent MSRVs.
- dovi_tool supported DV profiles: 4, 5, 7, 8.1, 8.4. Profile 8.2 is NOT documented as supported — if a user asks, do not assume it works. Profile 7 is BL+EL (dual-layer) and can be
convert-ed to 8.1 (single-layer with RPU).
- Canonical dovi_tool subcommands (verified):
info, generate, editor, export, plot, convert, demux, mux, extract-rpu, inject-rpu, remove.
- Canonical hdr10plus_tool subcommands (verified):
extract, inject, remove, plot, editor. No convert / demux / mux — those are dovi_tool only.
- Input to
extract-rpu and extract must be HEVC Annex-B bytestream, NOT MP4/MKV-muxed HEVC. Extract it first: ffmpeg -i in.mkv -c:v copy -bsf:v hevc_mp4toannexb -f hevc - | dovi_tool extract-rpu -.
- After
inject-rpu / inject, you get Annex-B HEVC again — you have to re-mux into MKV via mkvmerge or MP4 via MP4Box. ffmpeg's MP4 muxer for HEVC+RPU is flaky; MP4Box is recommended.
- GitHub API rate limit: unauthenticated requests are capped at 60/hour. The release-info pages use
api.github.com. Set GITHUB_TOKEN=… env var for 5000/hour if you hit the limit.
- Release-note pages are JSON from the GitHub API — text-extracted to show tag name, published date, and body.
- Docs are Markdown from
raw.githubusercontent.com — preserved as-is by the script (no HTML→text conversion). Headings with ## map cleanly to section --id Heading.
- Neither tool writes HDR10+ or Dolby Vision from scratch. They operate on existing RPU/SEI data. "Generate" in dovi_tool produces profile-compliant RPU from input like L1 analysis + Dolby XML — it's authoring, not AI. No tool "adds Dolby Vision to an SDR video" on its own.
Examples
Example 1 — "What Dolby Vision profiles does dovi_tool support?"
uv run ${CLAUDE_SKILL_DIR}/scripts/hdrdocs.py search --query "profile" --page dovi-readme --limit 10
Example 2 — "How does dovi_tool editor work?"
uv run ${CLAUDE_SKILL_DIR}/scripts/hdrdocs.py search --query "editor" --page dovi-readme
uv run ${CLAUDE_SKILL_DIR}/scripts/hdrdocs.py section --page dovi-readme --id editor
Example 3 — "What's the latest hdr10plus_tool version?"
uv run ${CLAUDE_SKILL_DIR}/scripts/hdrdocs.py fetch --page hdr10plus-releases
Example 4 — "How do I extract RPU from a Blu-ray remux?"
uv run ${CLAUDE_SKILL_DIR}/scripts/hdrdocs.py search --query "extract-rpu" --page dovi-readme
Example 5 — "Does hdr10plus_tool have a convert command?"
uv run ${CLAUDE_SKILL_DIR}/scripts/hdrdocs.py search --query "convert" --page hdr10plus-readme
Troubleshooting
Error: urlopen error 403 from GitHub API
Cause: Rate-limited (60 unauth requests/hour).
Solution: Set GITHUB_TOKEN=... env var (a public-repo-read token is enough).
Search returns no hits for a subcommand you think exists
Cause: Wrong tool or wrong page. convert is dovi_tool only; editor is in both; x265-encode isn't a real subcommand (that's a convenience preset in hdr-hdr10plus-tool wrapper).
Solution: Drop --page to search both READMEs. Consult references/subcommands.md.
Cache is stale after upstream release
Solution: clear-cache then index.
Reference docs
- Full subcommand + flag catalog (verified against current READMEs) →
references/subcommands.md.
