writing-shape

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were identified in the instructions or configuration files.
  • [COMMAND_EXECUTION]: The skill does not involve shell commands or script execution. It explicitly sets 'disable-model-invocation: true' in its metadata, which prevents the model from calling external tools or functions during execution.
  • [DATA_EXFILTRATION]: There are no network requests or patterns indicating data exfiltration. File interactions are restricted to reading the user's 'pile' of raw material and appending the drafted article content to a local file, both of which are standard for the skill's stated purpose.
  • [PROMPT_INJECTION]: While the skill ingests untrusted 'raw material', the instructions focus on structured extraction and formatting within a human-in-the-loop workflow, effectively mitigating the risk of instructions in the data influencing the agent's core behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 04:25 PM
Security Audit — agent-trust-hub — writing-shape