The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the open web (Reddit, X, Google Search) to generate a research table, creating a surface for embedded instructions in the scraped data to influence the agent.
Ingestion points: Social media feeds (Reddit, X) and Google news results navigated to in Step 2 of SKILL.md.
Boundary markers: Absent. The skill lacks instructions to ignore or isolate potential commands or malicious prompts embedded in the scraped content.
Capability inventory: The skill uses live browsing tools (Chrome extension, Playwright) and facilitates tool chaining by suggesting transitions to other skills like post-writer or post-formatter using the scraped data.
Sanitization: No sanitization or validation of the ingested text is performed before it is processed and presented back to the user or passed to other skills.
[PROMPT_INJECTION]: The instructions use specific override keywords and directives to control and bypass standard agent interaction patterns.
Evidence: The use of ## CRITICAL: Auto-start on load and the instruction Do not summarise the research method in SKILL.md are designed to override the agent's default behavior and safety summarization protocols.

niche-research