Load for research tasks, architectural investigations, and vendor, library, or technology comparisons.
71%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Low
Low-risk findings worth noting
Fix and improve this skill with Tessl
tessl review fix ./skills/deep-research/SKILL.mdThe skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The required workflow runs Exa web search (api.exa.ai/search) and ingests returned arbitrary web page text, summaries, and highlights into the report the agent then reads, but the attacker can publish poison and cannot force delivery without the agent's query selecting it.
The skill fetches instructions or code from an external URL at runtime, and the fetched content directly controls the agent’s prompts or executes code. This dynamic dependency allows the external source to modify the agent’s behavior without any changes to the skill itself.
The only inbound fetch is the README setup step `kendex add vanillagreencom/kendex`, which pulls the skill's installed code from a GitHub repository; that repo is a stable, attributable third-party project with no first-party or enterprise provenance and no warning signs, so it is a common/reputable unofficial dependency (medium). The Exa `/search` call is outbound data, and the local scripts are not remote.
vanillagreencom/kendex
3a2f84c
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.