coding-prompt
v1.1.0Optimize and refine AI programming prompts with constraints, scenario focus, and validation to improve coding session instructions and prevent vague or incom...
⭐ 1· 101·0 current·0 all-time
byhanli@neuhanli
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The skill's name and description match the runtime instructions: it reviews and rewrites coding prompts, uses local reference documents, and monitors coding sessions. It does not request unrelated binaries, environment variables, or external service credentials.
Instruction Scope
All runtime actions are confined to reading the included reference files and the conversation context. Mode 1 explicitly loads local reference files for diagnosis; Mode 2 limits proactive alerts to two high-priority signals and forbids loading reference files. The skill does not instruct the agent to read system files, environment variables, or transmit data to external endpoints.
Install Mechanism
There is no install spec and no code files that would be written or executed. Being instruction-only, it has a minimal on-disk footprint (the provided reference files) and no download/install risk.
Credentials
The skill requests no environment variables, credentials, or config paths. The evolution protocol only permits appending to its own references/learnings.md file on an explicit 'update skill' trigger.
Persistence & Privilege
always:false (good). The skill is designed to remain 'active' within a conversation once the user enables it, and it can autonomously issue in-session alerts per its Mode 2 rules. This autonomous monitoring is expected for a prompt-coaching skill but is something users should be aware of. The only writable surface it declares is references/learnings.md (append-only) when explicitly triggered.
Assessment
This skill is an instruction-only tool for improving coding prompts and appears internally consistent. Before installing, consider: 1) it will read the included reference files when you ask for prompt optimization (Mode 1) and will monitor the ongoing conversation for two specific high-priority signals if you activate it (Mode 2); 2) it requests no external credentials or installs and does not reference system files; 3) when you say 'update skill' it may append to its own references/learnings.md — review the appended content periodically if you worry about accumulating data or undesired guidance. If you want to limit autonomous behavior, simply avoid activating the monitoring mode or do not invoke the 'activate' trigger phrases.Like a lobster shell, security has layers — review code before you run it.
latestvk97decs4k82vmd0ngst8w1mkps83p6jf
License
MIT-0
Free to use, modify, and redistribute. No attribution required.