The Agent Skills Directory

[SAFE]: The skill's primary function is to generate content plans and briefs, which is a low-risk text generation task involving standard prompt-to-text workflows.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources which could potentially contain malicious instructions, though this is part of its core functionality.
Ingestion points: Reads brand-profile.md, voice.md, and content-pillars from the local environment to inform the content strategy.
Boundary markers: No explicit delimiter usage is specified for the ingested brand files to separate instructions from data.
Capability inventory: The skill can trigger the scheduling-and-queue tool and other content generation skills (e.g., caption-writer) based on the plan.
Sanitization: Does not describe specific input validation, escaping, or filtering for the ingested brand profiles.
Note: The skill explicitly mandates user confirmation before any scheduling occurs, which serves as a critical human-in-the-loop security control and significantly reduces the risk of autonomous exploitation.
[COMMAND_EXECUTION]: No instances of shell command execution, subprocess calls, or system-level operations are present in the skill instructions or reference files.
[DATA_EXFILTRATION]: No network operations or data transmission to external domains were identified. The skill only references internal platform tools and local files.
[OBFUSCATION]: The content is provided in clear text with no evidence of encoding (e.g., Base64), zero-width characters, or hidden strings.

batch-content-plan