tdg-personal/research-ops
Evidence-first current-state research workflow for ECC. Use when the user wants fresh facts, comparisons, enrichment, or a recommendation built from current public evidence and any supplied local context.
55
69%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Low
Low-risk findings worth noting
Low
Low-risk findings.
1 low severity finding. Worth noting, but not necessarily harmful.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The 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.
Why it was flagged
The SKILL.md workflow explicitly directs the agent to use "exa-search" for "current-web discovery" and "deep-research" for "multi-source synthesis with citations" to research "current public information," which means the agent will fetch and read open/public third‑party web content (potentially user‑generated) and use it to drive recommendations and next actions.
Where we found it
third-party web content via exa-search
content-type · 3 sites
The plugin directs the agent to invoke `exa-search` for "fast current-web discovery," which fetches and returns open/public third-party web content that the agent then acts on, creating an indirect prompt-injection surface.
SKILL.md
17
`exa-search` for fast current-web discovery
SKILL.md
63
use `exa-search` for fast discovery
SKILL.md
11
it is not a replacement for `deep-research`, `exa-search`, or `market-research`
third-party web content via deep-research
content-type · 3 sites
The plugin directs the agent to invoke `deep-research` for "multi-source synthesis with citations," which fetches and synthesizes content from multiple third-party web sources, creating an indirect prompt-injection surface.
SKILL.md
18
`deep-research` for multi-source synthesis with citations
SKILL.md
64
escalate to `deep-research` when synthesis or multiple sources matter
SKILL.md
11
it is not a replacement for `deep-research`, `exa-search`, or `market-research`
third-party web content via market-research
content-type · 3 sites
The plugin directs the agent to invoke `market-research` for recommendations built from external research, which fetches and processes third-party web content, creating an indirect prompt-injection surface.
SKILL.md
19
`market-research` when the end result should be a recommendation or ranked decision
SKILL.md
65
use `market-research` when the outcome should end in a recommendation
SKILL.md
11
it is not a replacement for `deep-research`, `exa-search`, or `market-research`
third-party web content via lead-intelligence
content-type · 2 sites
The plugin directs the agent to invoke `lead-intelligence` for people/company targeting, which fetches third-party public information about individuals and companies, creating an indirect prompt-injection surface.
SKILL.md
20
`lead-intelligence` when the task is people/company targeting instead of generic research
SKILL.md
66
hand off to `lead-intelligence` when the real ask is target ranking or warm-path discovery
- Audited
- Security analysis