Henry Skill Installer
v1.0.0Scaffolds new skills safely in the workspace by creating skill folders, generating SKILL.md and minimal files, and verifying skill registration.
⭐ 0· 332·2 current·2 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The name/description claim to scaffold new skills in the workspace and the SKILL.md contains a step-by-step scaffolding workflow that fits that purpose. Minor mismatch: SKILL.md prefers using /usr/bin/tee and /bin/cat for writes/verifications but the skill declares no required binaries — this is a small documentation/metadata inconsistency that should be clarified.
Instruction Scope
Instructions limit writes to a single hard-coded path (/Users/clawdbot/.openclaw/workspace/skills/) and forbid deletions and network calls, which keeps scope narrow. However the line 'Never print tool JSON; call exec.' is ambiguous: it encourages executing commands rather than showing tool manifests and could give the agent broad discretion to run shell commands. Because the skill is instruction-only and relies on the agent to run commands (tee/cat/exec), that ambiguity increases risk and should be clarified.
Install Mechanism
There is no install spec and no code files — the skill is instruction-only. That minimizes the risk of arbitrary code being downloaded and executed at install time.
Credentials
The skill does not request environment variables, credentials, or config paths beyond the single workspace path it intends to write to. Required privileges appear proportionate to the stated goal.
Persistence & Privilege
always is false and the skill does not request persistent or cross-skill modifications. Autonomous invocation is allowed (platform default) but is not combined with broad credentials or persistent privileges here.
What to consider before installing
This skill is mostly coherent for local scaffolding, but take these precautions before installing: 1) Confirm the hard-coded workspace path (/Users/clawdbot/.openclaw/workspace/skills/) matches where you want files created — if not, request or edit the SKILL.md to accept a configurable path. 2) Ask the author to clarify what they mean by 'Never print tool JSON; call exec.' and to narrow which commands may be executed (avoid broad 'exec' instructions). 3) Note SKILL.md references /usr/bin/tee and /bin/cat — ensure those binaries are present and the skill metadata is updated to list them. 4) Run first-time use in a sandbox or with backups of the workspace to confirm generated files are safe. 5) If you need stronger guarantees, request the author include explicit checks (e.g., validate filenames, avoid interpolation of untrusted input) or require a preview step so you can inspect generated SKILL.md and files before the skill writes them.Like a lobster shell, security has layers — review code before you run it.
latestvk973pa9rgbpd1ndedz2wb8rqv5826h45
License
MIT-0
Free to use, modify, and redistribute. No attribution required.