Kol Content Screening
PassAudited by ClawScan on May 7, 2026.
Overview
This instruction-only skill is coherent for web-search-based KOL screening, but it will send account/keyword queries to search tools, use subagents for larger lists, and save report files.
This appears safe to install as an instruction-only research workflow. Before using it, make sure the KOL list, campaign keywords, and time window are okay to send through external web search, and review the generated markdown reports before forwarding them to clients or teammates.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Large KOL lists may trigger many automated search tasks, which can consume time, quota, or tool budget.
The skill intentionally fans out work to parallel subagents for larger account lists. This is consistent with the screening purpose and is bounded by group size, but it can increase tool usage and should be user-scoped.
For >15 accounts on one platform, split into groups of 8–10 and spawn parallel sub-agents.
Confirm the platforms, account list, keywords, time window, and desired output before starting large runs.
Confidential campaign lists or keywords could be exposed to the search provider or subagent workflow through search queries.
The workflow sends KOL nicknames, handles, and keywords to a web search provider and may delegate them to subagents. This is expected for web-search screening, but the provider is not fixed beyond 'or equivalent'.
Each sub-agent, for each account, runs at least two queries on the chosen web search tool (xiaosu-search or equivalent)
Only provide account lists and keywords that are acceptable to use with external search tools; avoid confidential campaign names unless disclosure to the search provider is acceptable.
The generated files may contain competitive-intelligence research and should be treated as sensitive if the account list or campaign topic is private.
The skill saves retrieved evidence, rankings, and user-supplied account data into local markdown files. This is useful and disclosed, but it creates persistent artifacts that may be read or reused later.
Output to `<workdir>/<keyword-slug>-kol-screening-{YYYYMMDD}.md` (markdown table) plus per-platform group files.Store the workdir in an appropriate location, review outputs before sharing, and delete report files when they are no longer needed.