Deslop the Copy
v1.0.0Strip AI-generated writing patterns and make text sound like a real person wrote it. Use this skill whenever the user asks to deslop, de-slop, clean up, huma...
⭐ 0· 66·0 current·0 all-time
byArtur Schaback@skyzer
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description match the actual instructions: the skill is an instruction-only editor that rewrites user-supplied text to remove AI 'tells' and humanize it. It does not request unrelated credentials, binaries, or system paths.
Instruction Scope
SKILL.md stays within the stated purpose (it operates on user-provided text) and does not instruct reading environment files or exfiltrating data. However, Pass 2 explicitly tells the agent to 'add opinions' and to vary rhythm aggressively — this can change facts or introduce invented assertions if the user expects pure copy-editing. Also the strict ban on em dashes is a global stylistic rule that could alter nuance.
Install Mechanism
The skill is instruction-only (no install spec, no code files), which is low-risk. The README suggests cloning a GitHub repo (https://github.com/skyzer/deslop-the-copy.git) as an installation option; that is outside the platform install metadata and would write code to disk if performed, so review the repository before cloning.
Credentials
No environment variables, credentials, or config paths are required. The skill does not request any external tokens, which is proportional to a text-rewriting utility.
Persistence & Privilege
No elevated persistence is requested (always: false). The skill does not modify other skills or request system-wide configuration changes. Autonomous invocation is allowed by default but not combined with other red flags.
Assessment
This skill is internally consistent and low-risk from an access/credential perspective: it only needs the text you give it and asks for no secrets. Before using or installing, consider: 1) It will not only edit phrasing but may add opinions or reframe factual statements — avoid using it on sensitive legal, medical, or financial copy unless you explicitly ask it to preserve factual accuracy. 2) The README suggests cloning a GitHub repo; if you follow that step, inspect the repository contents before running any code. 3) If you want strictly neutral edits, tell the skill not to invent opinions or to preserve facts; otherwise expect it to prioritize a human voice over literal fidelity. Finally, if you plan to let the agent invoke skills autonomously, be aware this skill can be called automatically but it has no special privileges or credential access.Like a lobster shell, security has layers — review code before you run it.
latestvk97eqdskb3x76j4189g2h0r09983tvwswritingvk97eqdskb3x76j4189g2h0r09983tvws
License
MIT-0
Free to use, modify, and redistribute. No attribution required.