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.
npx skills add lawchat-oss/taiwan-legal-plugin --skill cold-start-interviewgit clone --depth 1 https://github.com/lawchat-oss/taiwan-legal-pluginWrote 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.
[](https://agentmods.dev/skills/lawchat-oss/taiwan-legal-plugin/cold-start-interview)<a href="https://agentmods.dev/skills/lawchat-oss/taiwan-legal-plugin/cold-start-interview"><img src="https://agentmods.dev/badge/skills/lawchat-oss/taiwan-legal-plugin/cold-start-interview/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.
<a href="https://agentmods.dev/skills/lawchat-oss/taiwan-legal-plugin/cold-start-interview"><img src="https://agentmods.dev/badge/skills/lawchat-oss/taiwan-legal-plugin/cold-start-interview.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00091 | $0.01216 |
| Opus 5.5 | $0.00036 | $0.00486 |
| Sonnet 5.5 | $0.00018 | $0.00243 |
| Haiku 4.5 | $0.00009 | $0.00122 |
Grade A, and why
cold-start-interview 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 1mo ago.
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.
How it starts
The opening of the file, as written. The whole thing — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/taiwan-legal:cold-start-interview
Captures the user's research defaults and writes them to a practice profile.
Audience
The same audience as the plugin's lookup skills — Taiwan-legal researchers, attorneys, paralegals, law students. This is configuration only; no legal interpretation occurs here.
Output file
Writes ~/.claude/plugins/config/taiwan-legal-plugin/taiwan-legal/CLAUDE.md with this structure:
# taiwan-legal practice profile
## Judgment search defaults
- court_levels: [<list>]
- divisions: [<list>]
- default_date_window: <string>
## Statute lookup defaults
- exclude_abolished: <true | false>
## Citation style
- style: <string>
## Standing scope
- subjects: <free text>
Instructions
Ask the user the following, one at a time. After collecting answers, write the profile file. If the file already exists, ask whether to merge or replace before overwriting.
-
Court levels — Which Taiwan court levels do you most often research?
- Common picks: 最高法院 / 高等法院 / 地方法院 / 行政法院 / 智慧財產及商業法院
- Default if skipped: 最高法院、高等法院、地方法院
-
Divisions — Civil (民事), criminal (刑事), administrative (行政) — which do you focus on?
- Default if skipped: all three
-
Date window — Default search range when the user does not specify? (e.g., "last 5 years", "last 10 years", "no default")
- Default if skipped: last 5 years
-
Citation style — How should the plugin cite cases and statutes in summaries?
- Options: 完整版 (字號 + 法院 + 日期 + URL) / 簡潔版 (字號 + 條號) / Bluebook-style with Taiwan adaptations
- Default if skipped: 完整版
-
Abolished regulations — When searching regulations, exclude abolished ones by default?
- Default if skipped: true
-
Standing scope — Any standing subject focus that should be remembered for every query? (Optional; leave blank for none.)
Worked example
Input (user runs the skill):
/taiwan-legal:cold-start-interview
Interaction:
Q1: Which Taiwan court levels do you most often research? (default: 最高法院、高等法院、地方法院) User: 最高法院、智慧財產及商業法院 Q2: Civil / criminal / administrative — which? (default: all three) User: 民事、行政 Q3: Default date window? (default: last 5 years) User: last 10 years Q4: Citation style? (default: 完整版) User: Q5: Exclude abolished regulations by default? (default: true) User: Q6: Standing subject focus? (optional) User: 智慧財產
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.
- 1mo ago Changed · -1 tokens per session a902b5dd5818
- 2mo ago First seen · 131 lines · 92 tokens per session scan A d50e555c9f5c
cold-start-interview is a skill published in the GitHub repository lawchat-oss/taiwan-legal-plugin (43 stars, last pushed 18d ago), licensed MIT. It adds 91 tokens to every session and 1,216 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-08-30.
cold-start-interview compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| cold-start-interviewlawchat-oss/taiwan-legal-plugin | Skill | 43 | 18d ago | MIT | A | 91 | Claude Code |
| fto-triageanthropics/claude-for-legal | Skill | 9.9k | yesterday | Apache-2.0 | A | 81 | Claude Code |
| legal-holdanthropics/claude-for-legal | Skill | 9.9k | yesterday | Apache-2.0 | A | 68 | Claude Code |
| specification-writingwanshuiyin/Auto-claude-code-research-in-sleep | Skill | 19k | today | MIT | A | 49 | Claude Code |
| regulatory-research-fallbackHKUDS/OpenSpace | Skill | 7.9k | 2d ago | MIT | A | 20 | Claude Code, Codex |
| terms-of-servicegenspark-ai/genoffice | Skill | 12k | today | Apache-2.0 | A | 32 | Claude Code, Codex |
| x-scorecardx-cmd/x-cmd | Skill | 4.7k | 8d ago | Apache-2.0 | A | 57 | Claude Code, Codex |