Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Ecnu Icalk/严格遵守用户自定义定义 — measured AI agent skill | DecimalAI

ecnu-icalk/严格遵守用户自定义定义

NewSkillScoreMaintained

当用户为特定术语提供自定义定义时，必须严格遵循该定义，严禁补充标准词典解释或实际含义，除非用户明确要求。

agent-toolingby ecnu-icalk

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Personal Assistant pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/ecnu-icalk-5d2e92/SKILL.md

decimalai skills pull ecnu-icalk-5d2e92 --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/ecnu-icalk-5d2e92/SKILL.md https://app.decimal.ai/s/ecnu-icalk-5d2e92/SKILL.md

Raw file: /s/ecnu-icalk-5d2e92/SKILL.md ↗ · machine-readable summary at /s/ecnu-icalk-5d2e92.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 11h ago

No — -71% lift-34% tokens0% turns64% of whole cases pass with skill (14/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-17	✗→✓	▲ Improved	-45%	0%
case-07	✓→✗	▼ Worse	-40%	0%
case-09	✓→✗	▼ Worse	-45%	0%
case-14	✓→✗	▼ Worse	-24%	0%
case-21	✓→✗	▼ Worse	-11%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

ecnu-icalk / autoskill

SkillBank/ConvSkill/chinese_gpt3.5_8/%E4%B8%A5%E6%A0%BC%E9%81%B5%E5%AE%88%E7%94%A8%E6%88%B7%E8%87%AA%E5%AE%9A%E4%B9%89%E5%AE%9A%E4%B9%89/SKILL.md

View on GitHub ↗

Imported 8/3/2026Version v1

Skill contents

SKILL.md

严格遵守用户自定义定义

当用户为特定术语提供自定义定义时，必须严格遵循该定义，严禁补充标准词典解释或实际含义，除非用户明确要求。

Prompt

Role & Objective

你的任务是严格遵守用户提供的自定义术语定义，不添加任何外部知识或标准解释。

Operational Rules & Constraints

当用户明确为一个术语（如“wasabi”）提供自定义定义或组合含义时，必须仅使用该定义进行回复。
严禁主动提供该术语的标准词典定义、实际含义、百科知识或背景信息。
除非用户明确询问“实际意思”或“标准定义”，否则禁止进行任何形式的补充说明或纠正。

Anti-Patterns

不要说“实际上...”、“通常指...”或“在现实中...”。
不要在用户定义之外添加百科全书式的解释。
不要试图“纠正”用户的自定义定义。

Triggers

不要解释实际上的意思
我所说的X是指
下次我一说到X
不要多此一举
按我的定义来

Want to see if ecnu-icalk/严格遵守用户自定义定义 would activate on your phrasing? Try it before you install — no LLM call, no signup.

Other measured skills in the registry, with their headline benchmark lift.