skilld
All skills
aktsmm avatar

/book-writing-workspace

@b7ad2f8
by yamapanaktsmm/agent-skills26 stars
4

Operate a reusable technical book manuscript workspace with writing structure, reader persona SSOT, review rules, and optional Markdown to Re:VIEW/PDF support. Use when organizing a book manuscript repo, defining or reviewing target readers, standardizing chapter/section files, setting writing/review agents, handling publisher proof comments, or assessing an existing writing workspace. Triggers on book writing workspace, reader persona, reader audience, 読者ペルソナ, persona SSOT, technical book project, 執筆ワークスペース, manuscript workflow, Re:VIEW workspace, proof PDF, annotated proof, publisher proof comments, 校正, 校正注釈, and 初校.

Ask your Agent

Use this Skill: https://skilld.dev/gh/aktsmm/agent-skills/book-writing-workspace

This session only. Nothing lands on disk.

templatesdocsrelease-readiness-record.md

≈3.3k tokens on demand. Your agent reads this file only when SKILL.md points to it.

Release Readiness Record

Create one copy of this record for each release candidate. Keep reusable workflow rules outside this file.

Candidate

Field Value
Title {{BOOK_TITLE}}
Candidate ID YYYY-MM-DD-release-name
Source revision TODO
Manuscript SSOT path / identifier TODO
Candidate fingerprint TODO
Proof revision / identifier NOT_APPLICABLE
Proof artifact path / identifier NOT_APPLICABLE
Accepted baseline TODO
Change window TODO
Required deliverables TODO
Reviewer TODO
Release approver TODO
Exception approver TODO
Tie-break authority TODO
Authorship / originality basis TODO
Publication rights / contributor agreements TODO
Unattributed reused content search / disposition TODO

Enabled Modules

Use ENABLED or NOT_APPLICABLE. Every row requires a reason, decision owner, and evidence; blank dispositions block Gate 0.

Module Disposition Reason / acceptance standard Decision owner Evidence
Figures / screenshots ENABLED / NOT_APPLICABLE
Citations / external links ENABLED / NOT_APPLICABLE
Exercises / answer keys ENABLED / NOT_APPLICABLE
Glossary / index ENABLED / NOT_APPLICABLE
Runnable code / procedures ENABLED / NOT_APPLICABLE
Companion files ENABLED / NOT_APPLICABLE
Localization ENABLED / NOT_APPLICABLE
AI-generated / third-party assets ENABLED / NOT_APPLICABLE
Preview / promotional materials ENABLED / NOT_APPLICABLE
Accessibility ENABLED / NOT_APPLICABLE Standard / scope:
Privacy / consent ENABLED / NOT_APPLICABLE
Contractual approvals ENABLED / NOT_APPLICABLE
Disclaimers / regulated content ENABLED / NOT_APPLICABLE
Trade / territory restrictions ENABLED / NOT_APPLICABLE
Print / fixed PDF ENABLED / NOT_APPLICABLE
Reflowable EPUB ENABLED / NOT_APPLICABLE
Web ENABLED / NOT_APPLICABLE
Download package ENABLED / NOT_APPLICABLE
Retailer / distributor submission ENABLED / NOT_APPLICABLE
Release communication ENABLED / NOT_APPLICABLE
Backup / recovery ENABLED / NOT_APPLICABLE
Errata / support ENABLED / NOT_APPLICABLE
Revision / withdrawal ENABLED / NOT_APPLICABLE

Gate Results

Use PASS, FAIL, BLOCKED, PASS_WITH_EXCEPTIONS, or NOT_APPLICABLE.

Gate Status Method / command Scope / artifact Evidence Owner / next condition
0. Scope and evidence
1. Inventory preflight
2. Deterministic source checks
3. References and links
4. Content and reader journey
5. Figures, screenshots, and rights
6. Editorial and proof closure
7. Build, artifact, and channel review
8. Independent sign-off and manifest lock

Review Lanes

Lane ID Scope / expected count Candidate / source fingerprint Executor / model Reviewed count Result artifact Status / owner

Worker Result Acceptance

Allowed lane status values are COMPLETED, BLOCKED, or FAILED.

Lane ID Status valid Target match Count match Evidence fresh Artifact verified Cleanup verified Required fields missing / invalid Accepted / reason

When a result is rejected, record every missing or invalid required field by name; do not write only schema mismatch or failed.

Lane Findings Index

Lane ID Finding ID Priority Source location Evidence reference Status Owner / resolution

Use VERIFIED, CONFLICT, or UNVERIFIED for Status. A result artifact supplements this index; it does not replace the required finding-level evidence.

Retry and Escalation

Lane ID Attempt count Recoverable reason Hard-gate reason Escalated at Last valid state / artifact

High-Risk Changes

Change ID Before / after Authoritative source / owner Rejected alternatives Blast-radius result Invalidated gates Independent review / verdict

Grounding and Conflicts

Use VERIFIED, CONFLICT, or UNVERIFIED. Model output alone is not evidence.

Claim / finding ID Status Source location / final URL Supporting excerpt / section Retrieved at Verified by Resolution / owner

Link Inventory

Use claim for a link that supports a factual statement or answer, and navigation for a reader path that needs no evidence excerpt.

Source location Kind Displayed URL Final URL Locale Retrieved at Result Claim ID Evidence excerpt / section Owner / status

Rights and Approvals

Item / location Asset or claim type Permission / consent basis Attribution / restriction Approver / evidence Status

Compatibility and Accessibility

Channel / target Standard or environment Validator / review method Result Exceptions / owner

Delivery and Security

Item / channel Check Result / receipt Owner Follow-up trigger
Dependency / vulnerability scan
Malware / package integrity scan
Authentication / authorization
Credential handling / rotation
Transport security / network destinations
Data collection / retention / deletion / third-party sharing
Legal deposit / registration
Production / fulfillment handoff
Channel-final purchase / download / access
Product description / preview / promotion / announcement

If a channel-final transaction test is omitted, record the reason, decision owner, alternative evidence, and reopen condition in the final row. Do not leave it blank or mark it passed without reader-facing evidence.

Post-Release Readiness

Path Owner Trigger Procedure / evidence Status
Reproduce build
Restore candidate
Release communication
Errata / accessibility report
Corrected edition
Suspend / withdraw
Security / privacy incident

Exceptions

ID Priority Decision Approver Expiry / reopen condition Evidence

Release Manifest

Field Value
Source revision TODO
Proof revision / identifier NOT_APPLICABLE
Enabled modules snapshot TODO
Build command / configuration fingerprint TODO
Tool versions TODO
Staged or upload-ready package TODO
Package hash TODO
Manifest comparison PASS / FAIL
Deliverable Size Hash Built from source revision Verified

Decision

  • Verdict: RELEASE / DO_NOT_RELEASE
  • Decided by:
  • Decided at:
  • Remaining risks:
  • Invalidated gates after the last fix: None / TODO

Source: SKILL.md on GitHub

Third-party checks

1 warning14d5 checks · Risk SAFE
  • Gen Agent Trust Hub14d

    The skill provides a comprehensive workspace for technical book writing, including automation for manuscript management and PDF generation. All identified behaviors, such as Docker-based builds and external tool dependencies, are standard components of the intended technical publishing workflow and show no signs of malicious intent.

  • Socket14d

    No alerts

  • Snyk14d

    Risk: LOW · No issues

  • Runlayer6mo

    8/48 files flagged

  • ZeroLeaks5mo

    Score: 93/100 · 2 sections analyzed

Provenance

Signed by skilld at b7ad2f8. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 18 hours ago.

Activeupdated last week

Capability

argument-hint
本のテーマ、原稿リポジトリ名、使いたい執筆フロー
user-invocable
true
metadata
{
  "author": "yamapan (https://github.com/aktsmm)"
}

README badge

README badge for aktsmm/agent-skills/book-writing-workspace