Checker
v1.0.0QA and peer review agent. Reviews output from other agents before delivery. Use when: (1) content from Scribe needs review before posting, (2) research from...
⭐ 0· 380·0 current·0 all-time
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 instructions: the SKILL.md is a focused checklist and verdict workflow for reviewing outputs from other agents. There are no unrelated environment variables, binaries, or installs requested. One small mismatch: the skill declares no required config paths or integrations but the workflow mentions reporting results to "Cello" (an external recipient) without specifying how; that's an unresolved integration point.
Instruction Scope
Instructions stay within the stated purpose (review, checklist, produce a verdict). They do require writing review files to a hard-coded workspace path (/home/ubuntu/.openclaw/workspace/artifacts/checker/) and to "report verdict to Cello" — both actionable steps but not declared elsewhere. The SKILL.md also asks the reviewer to validate things like link validity and accuracy, which implies network access to verify URLs and possibly to check sources; that is consistent with review but not explicitly scoped or constrained.
Install Mechanism
Instruction-only skill with no install spec and no code files — minimal disk/write footprint beyond creating artifact files. This is the lowest-risk install posture.
Credentials
The skill requires no environment variables, credentials, or external packages. That is proportionate to its described role. Note: runtime activities implied by the checklist (e.g., checking links, validating sources, inspecting images) may cause the agent to perform network requests or access files passed in by other agents; those accesses are expected but the SKILL.md doesn't enumerate needed permissions.
Persistence & Privilege
always is false and model invocation is allowed (the platform default). The skill writes artifacts into its own workspace directory only; it does not request system-wide configuration changes or persistent privileges. No evidence of elevated or cross-skill configuration access.
Assessment
This skill looks like a straightforward QA/peer-review helper and is generally coherent and low-risk: it doesn't install anything or ask for credentials. Before installing, note two small gaps you may want clarified: (1) SKILL.md writes artifact files to a hard-coded path (/home/ubuntu/.openclaw/workspace/artifacts/checker/) — confirm that your agent runtime allows writing there and that you are comfortable with permitted files being stored in that location; (2) it instructs the agent to "report verdict to Cello" but does not define what Cello is or how to send the report (another skill, an API, a messaging endpoint?). Ask the skill author to clarify the reporting destination and any network permissions expected for link/source validation. If those are acceptable, the skill is reasonable to use.Like a lobster shell, security has layers — review code before you run it.
latestvk9793v1whmgfc4rp2cfd4t7scn81s8gt
License
MIT-0
Free to use, modify, and redistribute. No attribution required.