tweet-draft-reviewer

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes local bash scripts to automate the identification and filtering of tweet drafts.\n
  • Employs the find command to locate markdown files within the user-specified or default content/tweet-drafts directory.\n
  • Uses grep to verify file metadata for a 'reviewed: true' status to avoid redundant processing.\n
  • These commands are scoped to the local filesystem and serve the primary functionality of the skill.\n- [PROMPT_INJECTION]: The skill possesses an inherent surface for indirect prompt injection due to its core function of reading and analyzing external text files.\n
  • Ingestion points: The skill reads the contents of .md files from the local content/tweet-drafts folder (File: SKILL.md, SKILL-OC.md).\n
  • Boundary markers: There are no explicit delimiters or specific instructions for the agent to ignore instructions that might be embedded within the draft content.\n
  • Capability inventory: The skill allows the agent to execute shell commands for file discovery and filtering.\n
  • Sanitization: No content validation or sanitization is performed on the ingested draft text prior to evaluation against the voice rules.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 12:20 AM