The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill generates a temporary JavaScript module at runtime and executes it to bypass CLI argument limitations of the platform environment.
Evidence: In scripts/lib/launch.mjs, the runCli function constructs a JavaScript string using renderEntry, writes it to a temporary file (e.g., /tmp/ego-instagram-.../entry.mjs), and then executes it by spawning a subprocess with ego-browser nodejs -e "await import(...)".
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Instagram post captions and usernames, which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: Data is retrieved from the Instagram feed API and processed in scripts/lib/pure.mjs (via normalizeSavedItem) and scripts/lib/run.mjs.
Boundary markers: There are no delimiters or explicit instructions to the agent to ignore embedded commands within the fetched captions.
Capability inventory: The skill has capabilities to write files to the local disk (fs.writeFile) and make network requests to Instagram CDNs.
Sanitization: No sanitization, escaping, or filtering of the caption or username fields is performed before the data is processed or returned in the skill's summary output.

instagram-saved-download