Intention Engine
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This instruction-only skill is transparent about making the agent infer intent and sometimes push back, with no code, credentials, install hooks, or network behavior shown.
This skill appears safe as an instruction-only behavior guide, but it will make the agent more proactive about inferring your real goal and possibly pushing back. Use it if you want that style of assistance, and be clear about constraints or when you want literal execution.
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.
The agent may sometimes suggest a different path, push back, or ask clarifying questions instead of simply doing the literal request.
This instructs the agent to reason beyond the literal task and act toward inferred intent, which can change how user requests are handled.
When the user says "do A," A is one of many paths to the outcome they actually want. Your job is to understand the intention and execute toward it.
Install only if you want the agent to proactively infer intent; for important tasks, explicitly state constraints and ask the agent to confirm before changing course.
Past notes, memories, or project state may affect how the agent interprets current requests.
The skill relies on existing profile, memory, and project context to infer user intent, so inaccurate, stale, or sensitive context could influence the agent’s decisions.
Check Intention Sources... User profile goals — declared priorities (USER.md or equivalent)... Recent memory — last 2-3 days of decisions and conversation... Project/task state
Keep user profile and memory data accurate, review stale context periodically, and ask the agent to state what context it relied on for high-impact decisions.