Neckr0ik Freelance Automator
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This skill needs review because it advertises high-impact freelance account automation while authentication, approvals, persistence, and even real job-search behavior are not clearly bounded.
Before installing or using this skill, verify whether its job results are real, avoid connecting freelance accounts or credentials until scope is documented, and require manual review for every proposal, message, follow-up, and invoice. Be aware that it stores business data in ~/.freelance-automator and invokes a local Ollama model for proposal generation.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
You could believe the tool found real freelance opportunities when it may only be producing demo listings.
The script explicitly says it generates example jobs rather than searching real platforms, which conflicts with SKILL.md claims of automatic job hunting and full platform support.
# For now, generate realistic job examples ... # In real implementation, would search actual platforms
Treat job results as unverified; the publisher should clearly label demo behavior or implement and document real platform searches before claiming full support.
If the skill later asks for or uses platform account access, the required permissions and handling rules are not defined up front.
The code marks supported freelance platforms as authentication-required, while the registry metadata declares no primary credential, required env vars, or config paths.
"requires_auth": True
Do not provide platform credentials until the skill documents exactly which accounts, scopes, storage locations, and actions are required.
Automated messages could be sent to clients later or repeatedly without the user reviewing each message.
The skill advertises ongoing autonomous client communication without describing stop conditions, review requirements, recipient limits, or cancellation controls.
auto-reply Set up auto-reply for common questions follow-up Schedule follow-up messages
Use only draft/review modes unless the skill adds explicit confirmation, scheduling limits, logs, and an easy disable mechanism.
A wrong invoice amount, recipient, or project description could be sent to a real client.
Direct invoice emailing is purpose-aligned for a freelance tool, but it is a high-impact business action that should be reviewed before use.
--send Send directly to client email
Preview every invoice and require explicit human approval before using any direct-send option.
Client and proposal data may remain on the machine across sessions.
The script creates persistent local storage for jobs, proposals, and client records.
self.config_dir = Path(config_dir or Path.home() / ".freelance-automator") self.jobs_dir = self.config_dir / "jobs" self.proposals_dir = self.config_dir / "proposals" self.clients_dir = self.config_dir / "clients"
Review and periodically clean ~/.freelance-automator, and avoid storing sensitive client data unless retention and access expectations are acceptable.
Proposal content and job details are passed to a local model command, and the tool may fail or behave differently if Ollama is not installed.
The script runs a local Ollama model to generate proposals, but no required binary is declared in the metadata.
subprocess.run(["ollama", "run", "llama3.2:latest", prompt], capture_output=True, text=True, timeout=60)
Install and trust Ollama separately if you choose to use this feature, and confirm what data is included in prompts.