What is review-opinion-delivery?

A workflow for turning an architect's existing construction-drawing review notes into a Chinese Word document with linked evidence and checked pages.

Use it to package approved or draft DOCX opinions, marked DWG or PDF locations, and screenshots into a verified review report.

It keeps the reviewer's conclusions unchanged while organizing the supporting drawings, screenshots, and document details for delivery.

review-opinion-delivery — Skill for Codex by jack20040828

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add jack20040828/kangmaojian-skills --skill review-opinion-delivery

Clone the repo

git clone --depth 1 https://github.com/jack20040828/kangmaojian-skills

Made for: Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/jack20040828/kangmaojian-skills/review-opinion-delivery/github.svg)](https://agentmods.dev/skills/jack20040828/kangmaojian-skills/review-opinion-delivery)

Your own site

<a href="https://agentmods.dev/skills/jack20040828/kangmaojian-skills/review-opinion-delivery"><img src="https://agentmods.dev/badge/skills/jack20040828/kangmaojian-skills/review-opinion-delivery/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/jack20040828/kangmaojian-skills/review-opinion-delivery"><img src="https://agentmods.dev/badge/skills/jack20040828/kangmaojian-skills/review-opinion-delivery.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 150 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,234 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00150	$0.01234
Opus 5	$0.00075	$0.00617
Sonnet 5	$0.00030	$0.00247
Haiku 4.5	$0.00015	$0.00123

Measured 12d ago against content hash 93602d9a9615, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

review-opinion-delivery scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

The scan reads SKILL.md. This mod also ships 15 executable files (evals/run_regression_checks.py, evals/run_v12_regression_checks.py, scripts/compare_reviewer_docx.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

review-opinion-delivery/SKILL.md · 57 lines

How it starts

The opening of the file, as written. The whole thing — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.

施工图审查意见整理交付

Package opinions already made by the reviewer into a faithful, evidence-linked Word deliverable. Keep this workflow separate from technical drawing review.

Non-Negotiable Boundary

Treat the reviewer's latest confirmed DOCX as the highest content and style authority. Preserve its approved wording, item scope, certainty, regulation text, item consolidation, and mandatory/suggested status.
Do not add, remove, strengthen, weaken, merge, split, or infer a technical conclusion without explicit reviewer approval.
Use drawings only to locate marks and verify transcription. Do not derive new issues from them.
Preserve 无 only when it is reviewer-approved or explicitly confirmed as an internal contradiction, missing name/index, or design-expression issue that does not assert an external threshold. An unreviewed placeholder 无 blocks delivery; regulation research, completion, and conclusion reclassification require a separately authorized building-review phase.
Correct obvious typos, punctuation errors, and deterministic opinion-type aliases; record the before/after/reason in edit_log.csv.
If a correction has two plausible results or could change 必须修改/建议修改, technical meaning, applicability, or responsibility, mark it 需确认 and stop delivery.
Every opinion type must resolve to one of the 12 canonical values in references/opinion-types.json; invalid or unresolved values are a hard failure.
Record material source conflicts as 需确认; do not silently correct professional substance.

Read references/scope-boundary.md before editing any opinion.

Required Workflow

Create a workspace with scripts/create_delivery_workspace.py. For an existing project, pass --root <project>\03_审图过程.
Preserve every original under source/. Never overwrite a source file.
After the source set is stable, run scripts/snapshot_source_integrity.py <workspace>.
Build the v1.2 delivery_manifest.json with immutable item_id, display item_no, approved fields, provenance, evidence strategy, and zero or more evidence_images. Distinguish reviewer-approved regulation 无 from an unreviewed placeholder in the verification record. Fill verification_log.csv and edit_log.csv.
For marked DWG evidence, plot a working copy to PDF and crop only from rendered PDF pages. Never place CAD/T20 interface screenshots in the formal Word.
Run scripts/normalize_opinion_types.py <workspace> and then scripts/validate_delivery_package.py <workspace> before editing or generating the final Word. Deterministic corrections are mandatory; unresolved types must block delivery.
If a reviewer-approved DOCX exists, use the newest one as the base and make only approved local edits. Use scripts/generate_delivery_report.py only when no base DOCX exists.
Run scripts/validate_docx_content.py; formal Word text must not expose notes, note IDs, 得到大脑, evidence-source labels, PDF page traces, mark numbers, crop coordinates, or red-box explanations. Export to PDF, render every page, inspect every page, then run scripts/validate_report_qa.py.
Deliver only after both content and visual gates pass. Copy the final DOCX to <project>\04_审图成果; do not make scripts guess the project root.

Read the full file on GitHub · 57 lines

Files

What ships with it

26 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

12d ago First seen · 57 lines · 150 tokens per session scan A 93602d9a9615

Subscribe to this mod's changes

review-opinion-delivery is a skill published in the GitHub repository jack20040828/kangmaojian-skills (3 stars, last pushed 19d ago), licensed MIT. It adds 150 tokens to every session and 1,234 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Claude Code Documents Writing and docs agent-skills architecture codex construction-drawings document-automation

code-to-diagram

Analyze codebases and automatically generate architecture diagrams, flowcharts, and org charts. Uses AST parsing to map import dependencies for Python, JS/TS, Go, and Java, outputting Mermaid or SVG files. Triggered when users ask to visualize code architecture, understand dependencies, draw a flowchart, or create a…

zebbern/claude-code-guide · 74 tokens

review-opinion-delivery

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

施工图审查意见整理交付

Non-Negotiable Boundary

Required Workflow

What ships with it

What this file has done since we first saw it

Other skills, from other repositories

code-to-diagram

csv-to-sif

epd-parser

product-data-cleanup

product-data-import

product-spec-pdf-parser