The Agent Skills Directory

[SAFE]: No security issues were identified in the skill. The instructions are focused on providing technical specifications for social media publishing and rely on legitimate tool interactions.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface because it processes external brand profiles and reel assets. However, this is assessed as safe in this context because the skill's capabilities are restricted to generating media briefs and using structured publishing tools.
Ingestion points: brand-profile and finished Reel asset/topic are ingested into the agent context (referenced in SKILL.md and references/scope-and-connections.md).
Boundary markers: Absent; the skill does not instruct the agent to use specific delimiters or ignore instructions within these inputs.
Capability inventory: The agent can spec publish settings and call tools like WoopSocial for publishing or ideogram for design, but it has no arbitrary shell or unrestricted network access.
Sanitization: There is no evidence of input validation or sanitization before processing the external assets.

instagram-reels-publishing