以文言文压缩输出、节省 token。当用户要求文言回复、节省 token、省输出时触发。支持三档：极致(/e/Extreme)、适度(/m/Medium)、轻微(/l/Light)。
v1.0.0以文言文压缩输出、节省 token。当用户要求文言回复、节省 token、省输出时触发。支持三档：极致(/e/Extreme)、适度(/m/Medium)、轻微(/l/Light)。
⭐ 0· 93·0 current·0 all-time
by@u5bf
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name, description, and required artifacts match: the skill is an instruction-only compressor to produce classical-Chinese outputs and only references an included rules file. It declares no binaries, env vars, or other resources that would be unexpected for this purpose.
Instruction Scope
SKILL.md gives concrete conversion rules, preserves code/paths/versions, and asks the agent to report compression stats. The only caution: the skill explicitly makes 'token saving' the primary goal and permits aggressive, lossy compression (especially at /e). That is coherent with the stated purpose but may remove nuance or clarity for sensitive/technical explanations despite the stated preservation rules—test on representative content to ensure important details are not lost.
Install Mechanism
No install spec and no code files that run on the host; the skill is instruction-only, so nothing is written to disk or downloaded at install time.
Credentials
The skill requests no environment variables, credentials, or config paths—proportional and minimal for an output-formatting instruction set.
Persistence & Privilege
always is false and the skill is user-invocable. It does not request persistent presence or modify other skills or system settings.
Assessment
This skill is internally consistent and low-risk: it only contains rules for compressing text into classical Chinese and does not access credentials or external systems. Before enabling broadly, try several representative prompts (including technical instructions, error messages, and code) to confirm the chosen compression level preserves the details you need—especially for the /e extreme mode which purposely sacrifices readability for token savings. Avoid using it on sensitive legal/medical/security instructions without manual review. If you rely on fully faithful technical output, prefer /m or /l and verify the '[原~N字→文~M字，省X%]' summary the skill appends.Like a lobster shell, security has layers — review code before you run it.
latestvk97d5ehdexhrdrjfz03tnh48qn84gz4a
License
MIT-0
Free to use, modify, and redistribute. No attribution required.