simulated-upload-helper

Registry description claims the skill will "automatically create and upload a simple test skill" to demonstrate creation+upload. The SKILL.md, however, explicitly forbids creating new skill directories or SKILL.md files and says it should only trigger existing publish scripts. This is a direct contradiction: either the skill creates test skills or it does not. The source and homepage are also unknown which reduces confidence in intent.

SKILL.md instructs the agent to run local publish scripts (./publish-skills.sh --root ".cursor/skills" or with --skill "<dir>"). Running repository publish scripts is within the declared purpose (publishing skills) but SKILL.md has no safety checks: it doesn't verify what the publish script will do, doesn't require explicit user confirmation before publishing, and only 'forbids' creation/modification in prose which may be violated by the script itself. The instructions assume a specific repo layout (.cursor/skills) and permit non-interactive login via CLAWHUB_TOKEN, which enables network operations using credentials. These runtime actions have real side effects (network uploads, registry writes) and therefore require careful review of the publish script and explicit user authorization before execution.

Instruction-only skill with no install spec and no code files. This is low-risk from an installation perspective (nothing is written to disk by an installer).

Registry metadata declares no required environment variables, but SKILL.md references two optional environment variables (CLAWHUB_TOKEN and CLAWHUB_REGISTRY). Asking for CLAWHUB_TOKEN is reasonable for non-interactive registry uploads, but the metadata should declare it as a credential. The mismatch means the skill may prompt for or rely on sensitive tokens not declared in the registry metadata. Users should be cautious about providing CLAWHUB_TOKEN and verify why it's needed and how it's used.

always is false and there are no requested config paths or persistent installation steps. The skill does not request elevated platform privileges in its metadata. Autonomous invocation is allowed (platform default), which is expected for interactive skills; combine this with the instruction scope concerns before enabling autonomous runs.