Plea
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This instruction-only legal skill has no code, but it tells an agent to broadly scan private data and autonomously make or submit high-impact legal responses without clear user approval or boundaries.
Install only if you can keep it strictly draft-only. Before using it, set firm limits on which files or messages it may read, require explicit confirmation before any external submission, and have a qualified lawyer review all legal outputs.
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.
The agent could inspect or index sensitive emails, logs, and transaction records beyond what the user intended for a specific legal draft.
This describes broad access to private communications and records, with no stated path limits, user selection, exclusions, retention policy, or controls on reuse.
Fact-Finding: The Plea agent scans all relevant logs, emails, and transaction hashes.
Limit use to user-selected documents, require explicit consent before reading private files or mail, and avoid retaining or reusing indexed material unless the user approves.
A mistaken or premature filing could create real legal consequences, missed defenses, public records, fees, or procedural harm.
Court e-filing is a high-impact external action; the artifacts do not state that filings must remain drafts, require explicit approval, or be reviewed before submission.
"filing": "Standardized E-filing protocol integration for instant judicial submission"
Make the skill draft-only by default and require clear user approval plus qualified legal review before any submission to a court, arbitrator, or public system.
The agent may act as though it should respond to legal service or deadlines automatically, which can be unsafe if facts, jurisdiction, or legal strategy are wrong.
This implies autonomous event-triggered legal response behavior with no clear user control, monitoring scope, or confirmation step.
Plea automatically generates the initial Response to Complaint, preventing "Default Judgments" by ensuring the system always answers within the statutory window.
Require the user to invoke the skill for a specific matter and approve every generated response; do not allow background monitoring or automatic legal actions.
Users may rely on unsupported assurances about court recognition, legal quality, or default-judgment prevention in situations where professional legal advice is needed.
The skill makes strong legal-effectiveness claims without evidence or caveats, which could lead users to over-trust generated pleadings or filings.
delivered in a manner that the physical court system (or digital arbitration) must recognize
Treat outputs as unverified drafts only, add clear disclaimers, and advise users to consult a licensed attorney before using any document in a real proceeding.