calibre

name	calibre
description	Use when the user wants to search, retrieve, send, add, update, sync, export, convert, or clean books from the managed Calibre library runtime.
metadata	{"short-description":"Calibre library management via managed runtime"}

Calibre

Windows Runtime Commands

On native Windows, use the managed Windows runner and the native runtime command target. Set $runtime to the installed runtime root. Multi-agent installs usually use %LOCALAPPDATA%\ai-agents-skills\runtime. Then run:

$runtime = if ($env:AAS_RUNTIME_ROOT) { $env:AAS_RUNTIME_ROOT } else { "$env:LOCALAPPDATA\ai-agents-skills\runtime" }
& "$runtime\run_skill.bat" "skills/calibre/run_cal.bat" <args>

POSIX examples below use run_skill.sh and .sh command targets; use the Windows command target above on native Windows.

This uses the managed ai-agents-skills runtime copy of the Calibre workflow.

When to use

• search the Calibre library
• retrieve or send an ebook
• retrieve by book ID or disambiguation index
• add a new book file
• update book metadata or tags
• add or remove tags, and list shelves
• sync or doctor the library
• remove a book with dry-run support
• export metadata or convert formats
• clean staging files

Routing boundary

• Prefer this skill for explicit Calibre library operations and ebook workflows.
• Do not use this in place of zotero for generic "find/get/share/download a paper, DOI, ISBN, or book" requests; the top-level routing rule handles those with Zotero first.
• For review tasks that require locating a paper or book and the user did not supply the file/path, use Calibre immediately after Zotero and before any online retrieval.
• If Zotero does not satisfy a generic retrieval request, use getscipapers-requester only after the user explicitly says not to check/use the library or confirms outside retrieval after the library-first result is reported. Return to calibre only if the resulting file should be added to the ebook library.

Base path

• $AAS_RUNTIME_WORKSPACE/skills/calibre/

Use the managed runtime runner rather than invoking run_cal.sh directly.

Shared runner:

• bash "$AAS_RUNTIME_ROOT/run_skill.sh"

Local Library Profile Gate

Do not assume Calibre settings, library, or metadata.db locations. Before library-changing work, run or rely on a profile-aware read-only audit from the canonical installer:

cd ~/ai-agents-skills && make library-profile-audit ARGS="--profile library --json"

Discovery only lists candidates. It must distinguish authoritative Calibre libraries from runtime caches before any add, update, remove, convert, or sync operation. If no authoritative local Calibre database is found, mark the profile local-db-missing; do not create a library and do not treat cache DBs as writable libraries.

Supported system profiles:

• linux-local
• windows-mounted for Linux-side inspection of /windows/Users/...
• windows-native for native Windows execution

Calibre candidate validation must check metadata.db, quick-check status, book count, author/book file-tree consistency, canonical real path, symlink/mount or cloud-backed classification, and runtime-cache roots. Runtime caches are never mutation-authoritative even when their book count matches the real library.

Writes prefer a detected calibredb or calibredb.exe backend with an explicit library path. Guarded direct SQLite is fallback only and requires backup, lock, selected library root, and explicit warning. Windows-mounted or cloud-backed Calibre libraries are read-only from Linux unless the profile explicitly opts in after dry-run review.

Core commands

Use functions.exec_command.

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh search "<query>" [--format epub] [--tag fiction] [--limit 50] [--series "Series Name"]

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh get "<query>" [--format pdf] [--send "telegram:CHAT_ID"]

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh get --id 42 [--send "zulip:Research:books"]

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh get "ring" --index 0 [--send "telegram:CHAT_ID"]

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh add /path/book.epub [--isbn 9780140449136] [--title "X" --author "Y"] [--dry-run]

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh update --id 42 --title "X" --author "Y" --tags "a,b" --year 1965 --publisher "P" [--series "S" --series-index 1 --isbn 9780441013593]

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh add-tag --id 42 --tag "to-read"

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh remove-tag --id 42 --tag "to-read"

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh list-shelves [--tags|--series|--publishers]

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh sync [--force] [--progress]

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh remove "query" [--dry-run]

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh remove --id 42 [--dry-run]

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh convert --id 42 --to epub [--from pdf]

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh export --id 42 [--format bibtex]

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh doctor

bash "$AAS_RUNTIME_ROOT/run_skill.sh" skills/calibre/run_cal.sh clean

Important behaviors

• If get returns multiple matches, show candidates and ask the user to pick instead of guessing.
• Use --index only after showing candidates and receiving the user's selected index.
• Prefer dry-run before destructive operations like remove.
• For library-mutating or outward-facing actions (add, update, add-tag, remove-tag, remove, convert, sync, clean, get --send, or share), run the relevant dry-run/preview/search first when available, state the exact book/library/tag/channel to be affected, and get explicit confirmation unless the user's latest request already approved that exact action.
• Do not assume Calibre host dependencies such as ebook-convert are present; use doctor when conversion health matters.
• Sending books uses the OpenClaw file-sending path from the library workflow.
• --send uses channel:target syntax such as telegram:CHAT_ID, zulip:Stream:topic, googlechat:SPACE, or whatsapp:PHONE.
• add --isbn enriches metadata from Open Library before the library write.
• update --tags replaces the full tag set; use add-tag and remove-tag for incremental changes.
• Run sync at the start of a session if the library may have changed from Calibre desktop or another device.
• Use sync --progress when pulling metadata.db may take time. Progress is emitted as JSON lines on stderr so stdout remains the final JSON result.

Operational model

• Library access is profile-selected. Older Google-Drive-backed direct-SQLite cache workflows may still exist, but profile-aware local-library validation must run before treating any metadata.db as authoritative.
• Prefer a detected calibredb/calibredb.exe backend for authoritative library writes; direct SQLite is a guarded fallback only.
• Book files are downloaded to staging on demand rather than stored permanently in the workspace.
• After write operations, the updated metadata.db is pushed back to Drive and the local cache is refreshed.
• A file lock protects metadata.db from concurrent write conflicts.
• If Drive is unavailable, search can fall back to the last known local cache.
• clean removes staged files older than 24 hours.