Skill Creator
v0.1.0Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize a...
⭐ 0· 90·0 current·0 all-time
by@tlreal
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
Name and description match the SKILL.md instructions: the document is a guide for creating, evaluating, and iterating on skills. The skill requests no binaries, env vars, or installs, which is appropriate for an authoring/instruction-only tool.
Instruction Scope
The instructions are generally scoped to skill creation, interviews, test-case writing, and evaluation. However the SKILL.md explicitly references running 'claude-with-access-to-the-skill' and a script path 'eval-viewer/generate_review.py' and suggests using bundled 'scripts/'—none of these scripts or binaries are included in the skill bundle. This implies the agent may attempt to access or execute external files/tools in your environment or repo; verify where those live and whether you want the agent to run them. The instructions also advise reading conversation history and example files, which is expected for this purpose but means the agent may access user data if given permission.
Install Mechanism
No install spec and no code files are present, so nothing will be written to disk or auto-installed by the skill. This is low-risk from an install perspective.
Credentials
The skill declares no required environment variables, credentials, or config paths. SKILL.md mentions external tooling and scripts but does not require secrets or unrelated credentials in the metadata—this is proportionate to the described authoring purpose.
Persistence & Privilege
always:false and no special privileges are requested. The skill does not request to modify other skills or system-wide settings. Autonomous model invocation is allowed by default but is not a new permission introduced by this skill.
Assessment
This skill is an instruction-only guide for building and evaluating other skills and is coherent with its description. Before using it: (1) confirm where referenced scripts (e.g., eval-viewer/generate_review.py, any scripts/ directory) and the tool 'claude-with-access-to-the-skill' live—they are not bundled, so the agent would need permission to access/run files in your environment; (2) avoid providing secrets or credentials to the agent when running evaluations; (3) if you prefer to limit risk, run evaluation steps manually rather than allowing the agent to execute external scripts autonomously; (4) if you want higher assurance, ask the skill author for the missing scripts or a clear runbook showing exactly what will execute and what files will be read.Like a lobster shell, security has layers — review code before you run it.
latestvk97a79b6tyxd07kxdbw538zyd5835me7
License
MIT-0
Free to use, modify, and redistribute. No attribution required.