The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or credential exposure were detected. The skill consists of natural language instructions for content formatting and tone adaptation.
[SAFE]: No external downloads, package installations, or remote code execution patterns are present.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted data from various source versions (articles, threads, memos) to adapt them for social media.
Ingestion points: SKILL.md (Step 1: Selection of primary version from articles, memos, or changelogs)
Boundary markers: Absent; the instructions do not specify the use of delimiters or 'ignore' instructions for the source content.
Capability inventory: No local scripts or command execution; mentions related skills like x-api for publishing.
Sanitization: Absent; there is no validation or filtering specified for the ingested content.

crosspost