Deep Discussion
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This instruction-only skill is coherent for running multi-agent discussions, but users should expect multiple subagents and saved discussion logs.
This skill appears benign for structured multi-agent discussions. Use it when you are comfortable spawning several expert subagents and saving the full discussion record under workspace/deep-discussion/. Set clear limits on experts, agenda size, and runtime for large topics, and avoid putting secrets into the discussion unless you intend them to appear in the saved logs.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
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.
A discussion may launch several model subagents and continue through multiple agenda items, which can consume time and model usage.
The skill explicitly instructs the agent to spawn multiple expert subagents; this is central to the skill's purpose but affects runtime, cost, and task scope.
experts.forEach(id => sessions_spawn({ ... runtime: "subagent", thinking: "high" ... }))Before starting, confirm the number of experts, agenda size, and any desired stopping limits.
Sensitive information included in the topic or discussion may be visible to all spawned expert agents in the session.
The discussion protocol shares responses among spawned expert agents; this is expected for multi-agent discussion but means topic content is propagated across agent contexts.
You will see ALL other experts' responses
Avoid including secrets or private data unless you are comfortable with it being shared across the discussion agents.
The full discussion history may remain on disk after the session, including any sensitive details entered by the user.
The skill requires saving complete raw expert outputs to a persistent workspace log, which is aligned with the reporting purpose but creates retained context.
discussion-log.md | 完整讨论记录，包含所有专家的原始输出（必须！）
Review, move, or delete generated files after use if the discussion contains sensitive information.