Agent skill

Boundary Cases And Escalation

by h30190 in h30190/HJPLUS_Taiwan_Architect_KB

This skill should be used when a compliance question falls into a regulatory gray zone — the article uses discretionary wording (認定、核定、必要時、經主管機關認可), interpretations contradict each other…

CC-BY-SA-4.0Auto-check passed

Install Boundary Cases And Escalation

skills CLI
$ npx skills add h30190/HJPLUS_Taiwan_Architect_KB --skill boundary-cases-and-escalation -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install h30190/HJPLUS_Taiwan_Architect_KB boundary-cases-and-escalation --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/h30190/HJPLUS_Taiwan_Architect_KB.git skills-src && mkdir -p .claude/skills && cp -r skills-src/'raw/建築顧問方法論/邊界案例與函詢時機/boundary-cases-and-escalation' .claude/skills/boundary-cases-and-escalation && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
boundary-cases-and-escalation
GitHub stars
158
Token cost
~1.9k tokens
SKILL.md length
772 words
Files
1
Skills in repo
86
Repo updated
First seen
Licence
CC-BY-SA-4.0

At a glance

This skill should be used when a compliance question falls into a regulatory gray zone — the article uses discretionary wording (認定、核定、必要時、經主管機關認可), interpretations contradict each other…

  • Works in 4 steps: Search interpretations —… → Check local self-government rules and… → Check local 建管處 FAQ / pre-review… → …
  • SKILL.md covers Overview, Section 1: Boundary-Case…, Section 2: Known Gray Zones… and Section 3: The Escalation Ladder, plus 5 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Boundary Cases And Escalation is an agent skill from h30190/HJPLUS_Taiwan_Architect_KB. This skill should be used when a compliance question falls into a regulatory gray zone — the article uses discretionary wording (認定、核定、必要時、經主管機關認可), interpretations contradict each other, municipalities apply different review standards, or the measurement method is undefined (e.g., net depth of an irregular lobby). It teaches when to stop giving a definitive answer and recommend a formal inquiry (函詢) or pre-review (預審) to the competent authority instead.

Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts. Compatibility notes: claude-code,opencode,agent-skills

The repository describes itself as: 一個開源的台灣AEC產業知識庫,只要你願意共享經驗與知識就歡迎加入我們. The licence is CC-BY-SA-4.0.

Example prompts

  • “/boundary-cases-and-escalation”

Requirements

  • Python 3
  • Compatibility (from SKILL.md): claude-code,opencode,agent-skills

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. Search interpretations — taiwan-building-code_search_building_interpretations with the disputed noun + verbs like 認定/計算/適用. A…
  2. Check local self-government rules and review standards — the municipality may have published exactly the criterion the national code…
  3. Check local 建管處 FAQ / pre-review mechanisms — many municipalities publish Q&A or offer 建造執照預審 for contested design points.
  4. Recommend formal inquiry (函詢) or pre-review (預審) — when the determination is genuinely discretionary, the deliverable IS the…

What it can do on your machine

Read from SKILL.md and the folder at commit 35ed06e. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are python).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

  • Compatibility

    claude-code,opencode,agent-skills

    From compatibility in the SKILL.md frontmatter.

Context cost

Boundary Cases And Escalation loads about 1.9k tokens when it runs. Until then it costs about 122 tokens; SKILL.md has 772 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~122
When it runs · the whole SKILL.md, loaded when a task matches
~1.9k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from h30190/HJPLUS_Taiwan_Architect_KB at commit 35ed06e, republished under its CC-BY-SA-4.0 licence (© h30190). 772 words, ~1,856 tokens.

Download SKILL.mdSave it as .claude/skills/boundary-cases-and-escalation/SKILL.md (or your agent's skills folder).
name
boundary-cases-and-escalation
description
This skill should be used when a compliance question falls into a regulatory gray zone — the article uses discretionary wording (認定、核定、必要時、經主管機關認可), interpretations contradict each other, municipalities apply different review standards, or the measurement method is undefined (e.g., net depth of an irregular lobby). It teaches when to stop giving a definitive answer and recommend a formal inquiry (函詢) or pre-review (預審) to the competent authority instead.
compatibility
claude-code,opencode,agent-skills
type
Skill
user-invocable
true
license
CC-BY-SA-4.0
metadata.audience
architects
metadata.region
taiwan
metadata.class
C
metadata.status
verified
metadata.data-currency
2026-07-10

Boundary Cases & When to Escalate (邊界案例與函詢時機)

Overview

The single most important consulting judgment is knowing where determinable law ends and agency discretion begins. Under 建築法 §2, the competent building authority for permits is the municipal/county government — meaning that for many gray-zone questions, the legally correct answer is not a number but "this determination belongs to the local 建管處; inquire before filing."

An AI that guesses confidently in a gray zone can cause a client to design, file, and get rejected. This skill defines how to recognize a boundary case, the escalation ladder, and the required answer phrasing. It fills what was, before this skill, a complete gap in this knowledge base (zero occurrences of 函詢/建管處 escalation advice).


Section 1: Boundary-Case Recognition Signals

Treat the question as a boundary case when any of these appears:

#SignalExample
1Discretionary wording in the article「經主管機關認定」「必要時」「其他經主管機關認可之方式」
2Measurement method undefinedNet depth (淨深) of an irregular / flared lobby; where exactly to measure a projecting eave
3Interpretations conflict or are absentTwo 函釋 read differently; MCP interpretation search returns nothing on point
4Municipalities divergeSame article, but Taipei and Taichung review standards differ
5Aggregation/edge arithmeticA design sits exactly at a cap (10% / 15% / 2 m) or combines exempt items in an unregulated way
6Transitional/legacy status既存違建 treatment, existing lawful building undergoing partial alteration
7Classification edgeBuilding use group (使用類組) assignment where the actual use straddles two groups

Section 2: Known Gray Zones (seed list — extend as cases accumulate)

Gray zoneWhy it is grayRelated KB skill
FAR-exemption item boundaries(建築技術規則 §162)What counts as 機電設備空間/梯廳; caps interactfloor-area-exemption-pitfalls, balcony-lobby-far-recalculation
Lobby net-depth measurement on irregular plansNo statutory measurement diagram; local reviewers decidesame as above
Rooftop structures(屋突)height/area recognitionRecognition criteria applied variably—
Existing unauthorized construction(既存違建)Enforcement schedules differ per municipality—
Use-group assignment edges建築物使用類組及變更使用辦法 groups have overlapping real-world uses—
Open space / incentive FAR recognitionDiscretionary review committees—

Maintenance rule: when a consultation reveals a new gray zone (a question the ladder in Section 3 could not resolve), add it to this table with the outcome.

Section 3: The Escalation Ladder

Work down the ladder; each step is cheaper than the next. Only recommend step 4 when steps 1–3 fail to produce a determinate answer.

  1. Search interpretations — taiwan-building-code_search_building_interpretations with the disputed noun + verbs like 認定/計算/適用. A directly-on-point 函釋 usually resolves the case (cite it with agency + number + date).
  2. Check local self-government rules and review standards — the municipality may have published exactly the criterion the national code leaves open (審查基準、作業要點). Check the project municipality's 建管處 publications.
  3. Check local 建管處 FAQ / pre-review mechanisms — many municipalities publish Q&A or offer 建造執照預審 for contested design points.
  4. Recommend formal inquiry (函詢) or pre-review (預審) — when the determination is genuinely discretionary, the deliverable IS the recommendation to ask, plus a well-framed inquiry letter draft if helpful.
Show full SKILL.md (297 more words)Show less

Section 4: Required Answer Phrasing in a Gray Zone

Never present a gray-zone guess as a determinate answer. The professional format is: lean + basis + explicit locus of authority + recommended action.

Template (Chinese, for user-facing output):

text
依〔條文+函釋〕,本案傾向〔可/不可/可能〕……(理由:……)。
惟〔爭點,例如:不規則梯廳之淨深量測方式〕之認定權在〔○○市建管處〕,
目前無直接函釋可據。建議送件前以〔函詢/預審〕方式確認,避免退件重繪。

Rules:

  • State your lean and its legal basis — an escalation recommendation without analysis is lazy; analysis without the escalation caveat is dangerous.
  • Name the specific authority (which municipality's 建管處), not a vague "主管機關".
  • Quantify the downside of guessing wrong when possible (redesign, refile, schedule slip).
  • If drafting the inquiry letter, frame it around the specific article and the specific factual configuration — not "please advise generally".

Section 5: When NOT to Escalate

Escalation has costs (weeks of waiting). Do not recommend 函詢 when:

SituationCorrect action
The article is explicit and the facts fit squarelyAnswer with the article; cite it
A directly-on-point 函釋 existsAnswer with article + interpretation
Pure arithmetic under defined rulesCompute it (show the worked example)
The uncertainty is yours, not the law's (you haven't searched yet)Do steps 1–3 of the ladder first — do not outsource your homework to the agency

Section 6: AI Check Table

CheckConditionAI Action
Discretion ignoredAnswer states a determinate conclusion on a Section 1 signal without escalation caveatERROR: Rewrite per Section 4 template
Vague authorityAnswer says 「建議洽主管機關」 without naming which authorityWARNING: Name the specific municipality's 建管處
Premature escalation函詢 recommended before the ladder's steps 1–3 were attemptedWARNING: Search interpretations and local standards first
Gray zone not recordedA new gray zone surfaced but Section 2 table not updatedINFO: Propose adding it to the seed list

Section 7: MCP Tool Examples

python
# Step 1 of the ladder: is there an interpretation on point?
taiwan-building-code_search_building_interpretations(query="梯廳 淨深 不規則 認定")
taiwan-building-code_search_building_interpretations(query="屋突 高度 計算")

# Confirm the article's discretionary wording before classifying as gray zone
taiwan-building-code_search_building_code(query="經主管機關認定 容積", limit=10)

© h30190, CC-BY-SA-4.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in raw/建築顧問方法論/邊界案例與函詢時機/boundary-cases-and-escalation of h30190/HJPLUS_Taiwan_Architect_KB.

Open the folder on GitHubat commit 35ed06e

Compare with similar skills

Boundary Cases And Escalation next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Boundary Cases And Escalation compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Boundary Cases And Escalation this skillh30190/HJPLUS_Taiwan_Architect_KB158—~1.9kAutomated safety check: PassCC-BY-SA-4.0
Case Reviewsickn33/agentic-awesome-skills47k1 repos~1.6kAutomated safety check: PassMIT
Case Reviewzhaoxuya520/reverse-skill41k1 repos~1.6kAutomated safety check: WarnMIT
Clinical Case Reportnexu-io/open-design100k—~2.2kAutomated safety check: PassApache-2.0
Secops Casesgoogle/skills21k—~3.2kAutomated safety check: WarnApache-2.0
Building Soc Escalation Matrixmukul975/Anthropic-Cybersecurity-Skills34k—~1.8kAutomated safety check: PassApache-2.0

Similar skills

  • Case Review

    sickn33/agentic-awesome-skills

    Quality-gate review of a reverse-engineering or assessment case package: scope readiness, Evidence-to-Finding-to-Path traceability, work-item coverage, timeline consistency, and artifact hashes.

    47k GitHub starsUsed in 1 repo~1.6k tokens
    SecurityAuto-check passed
  • Case Review

    zhaoxuya520/reverse-skill

    Reviews a reverse-skill case package for scope readiness, Evidence to Finding to Path traceability, work item coverage, timeline references, and optional artifact hash integrity before report handoff.

    41k GitHub starsUsed in 1 repo~1.6k tokens
    SecurityAuto-check: warnings
  • Clinical Case Report

    nexu-io/open-design

    Structured medical case presentation for clinical rounds, conferences, and documentation.

    100k GitHub stars~2.2k tokensUpdated yesterday
    Auto-check passed
  • Secops Cases

    google/skills

    Official

    Manage Google Security Operations (SecOps) SOAR cases throughout their lifecycle.

    21k GitHub stars~3.2k tokensUpdated yesterday
    SecurityAuto-check: warnings
  • Building Soc Escalation Matrix

    mukul975/Anthropic-Cybersecurity-Skills

    Build a structured SOC escalation matrix defining severity tiers, response SLAs, tiered escalation paths, and notification procedures for security incidents, using context-driven criteria that…

    34k GitHub stars~1.8k tokensUpdated 1 mo ago
    SecurityAuto-check passed
  • Case Brief

    anthropics/claude-for-legal

    Official

    Brief a case in your preferred format. An agent skill from anthropics/claude-for-legal.

    9.6k GitHub starsUsed in 3 repos~1.6k tokens
    Auto-check passed

More from h30190/HJPLUS_Taiwan_Architect_KB

All 86 skills in this repo
  • Green Material Search Toolkit

    h30190/HJPLUS_Taiwan_Architect_KB

    This skill should be used when an architect or consultant needs to search Taiwan's TABC (財團法人臺灣建築中心) green building material certification database, assemble a set of qualified materials for a…

    158 GitHub stars~3.3k tokensUpdated 3 days ago
    Auto-check passed
  • Skill Name Hyphenated

    h30190/HJPLUS_Taiwan_Architect_KB

    This skill should be used when [specific trigger scenarios].

    158 GitHub stars~791 tokensUpdated 3 days ago
    Auto-check passed
  • Taichung Land Parcel Query

    h30190/HJPLUS_Taiwan_Architect_KB

    This skill should be used when an architect receives just an address or lot number (地號) for a site in Taichung City, Taiwan, and needs the land's basic data — zoning, registered land area, land…

    158 GitHub stars~1.9k tokensUpdated 3 days ago
    Auto-check passed
  • Permit Report Data Mapping

    h30190/HJPLUS_Taiwan_Architect_KB

    This skill should be used when mapping a legacy CPAMI permit report identifier to its actual BMS data groups, computed display fields, or unresolved template dependencies.

    158 GitHub stars~1.4k tokensUpdated 3 days ago
    Auto-check passed
  • Revit AI Targeting And Safe Execution

    h30190/HJPLUS_Taiwan_Architect_KB

    This skill should be used when an AI agent, MCP tool, Revit add-in, or automation must inspect or change a live Revit model and needs to prove its target set, choose bounded read-only context…

    158 GitHub stars~3.6k tokensUpdated 3 days ago
    Auto-check passed
  • Smoke Exhaust Review

    h30190/HJPLUS_Taiwan_Architect_KB

    This skill should be used when evaluating smoke exhaust compliance for buildings in Taiwan, including windowless floor determination, smoke compartment partitioning, smoke exhaust window effective…

    158 GitHub stars~1.9k tokensUpdated 3 days ago
    Auto-check passed

Questions about Boundary Cases And Escalation

What does Boundary Cases And Escalation do?

This skill should be used when a compliance question falls into a regulatory gray zone — the article uses discretionary wording (認定、核定、必要時、經主管機關認可), interpretations contradict each other…. Boundary Cases And Escalation is an agent skill from h30190/HJPLUS_Taiwan_Architect_KB., net depth of an irregular lobby).

How do I install Boundary Cases And Escalation in Claude Code?

Run `npx skills add h30190/HJPLUS_Taiwan_Architect_KB --skill boundary-cases-and-escalation -a claude-code`. Or copy the skill folder (raw/建築顧問方法論/邊界案例與函詢時機/boundary-cases-and-escalation in h30190/HJPLUS_Taiwan_Architect_KB) into .claude/skills/boundary-cases-and-escalation in your project. Claude Code loads it when a task matches its description.

How do I install Boundary Cases And Escalation in Codex?

Run `npx skills add h30190/HJPLUS_Taiwan_Architect_KB --skill boundary-cases-and-escalation -a codex`. Or copy the skill folder (raw/建築顧問方法論/邊界案例與函詢時機/boundary-cases-and-escalation in h30190/HJPLUS_Taiwan_Architect_KB) into .agents/skills/boundary-cases-and-escalation in your project. Codex loads it when a task matches its description.

Can I use Boundary Cases And Escalation in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add h30190/HJPLUS_Taiwan_Architect_KB --skill boundary-cases-and-escalation -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/boundary-cases-and-escalation, .gemini/skills/boundary-cases-and-escalation, .github/skills/boundary-cases-and-escalation and .opencode/skills/boundary-cases-and-escalation in your project.

What does Boundary Cases And Escalation need to run?

SKILL.md names no scripts, command-line tools or credentials: Boundary Cases And Escalation is instructions for the agent only. Our summary lists: Python 3. Compatibility (from SKILL.md): claude-code,opencode,agent-skills.

Does Boundary Cases And Escalation access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Boundary Cases And Escalation safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Boundary Cases And Escalation use?

Boundary Cases And Escalation is published under the CC-BY-SA-4.0 licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Boundary Cases And Escalation use?

About 1.9k tokens (SKILL.md is roughly 7.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Boundary Cases And Escalation?

Skills that share tags, products or a category with Boundary Cases And Escalation: Case Review (sickn33/agentic-awesome-skills, 47k stars), Case Review (zhaoxuya520/reverse-skill, 41k stars), Clinical Case Report (nexu-io/open-design, 100k stars) and Secops Cases (google/skills, 21k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Boundary Cases And Escalation?

h30190 (a GitHub user) maintains it in h30190/HJPLUS_Taiwan_Architect_KB, which has 158 GitHub stars. The repository holds 86 skills in this directory. The repository was last updated on October 8, 2026.

Source: h30190/HJPLUS_Taiwan_Architect_KB on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.