The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches transcription and metadata from api.supadata.ai. This is a targeted interaction with a data provider necessary for the skill's primary function.
[CREDENTIALS_UNSAFE]: Uses the $SUPADATA_API_KEY environment variable for authentication. No hardcoded secrets or sensitive credentials were found in the skill content.
[DATA_EXFILTRATION]: No unauthorized data transmission detected. Network activity is restricted to ingesting content from Instagram via the Supadata API and browser-use-cloud.
[DYNAMIC_EXECUTION]: The skill includes instructions to draft a new SKILL.md file if a reusable workflow is detected in the processed reel. This represents a controlled generation of markdown content from processed data.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Instagram Reel transcripts and captions.
Ingestion points: Untrusted data enters the context via the REEL_URL and subsequent transcript/caption extraction in SKILL.md.
Boundary markers: None identified; the agent is not explicitly instructed to ignore instructions embedded within the transcript.
Capability inventory: The skill has the capability to write files to the content-kb/ directory and draft new skill files.
Sanitization: No explicit sanitization or validation of the ingested text is performed before distillation. However, given the primary purpose of content ingestion, this is considered a low-risk profile.

laniameda-instagram-reel-digest