The Agent Skills Directory

[SAFE]: The skill operates within a restricted project scope, writing results to designated local directories and leveraging established CLI tools for all network and compute operations. No malicious patterns, obfuscation, or unauthorized access attempts were detected.
[INDIRECT_PROMPT_INJECTION]: As the skill's primary function is to ingest data from untrusted external community sources, it naturally possesses an indirect prompt injection surface. This risk is well-managed through the following controls:
Ingestion points: Untrusted data enters the context via the oma market harvest command (delegating to oma search fetch) from platforms such as Reddit, Hacker News, and Mastodon.
Boundary markers: Retrieved data is explicitly isolated into a 'Cluster Bank' during the intermediate processing stages before being synthesized by the LLM.
Capability inventory: The skill's capabilities are limited to reading from local cache/registry files and writing final markdown briefs to the .agents/results/market/ directory.
Sanitization: The oma market detect-trap pre-flight utility filters out ambiguous or personal-advice topics, while 'LAW' self-checks in the rendering stage validate the structural and content integrity of the final prose.

oma-market