Read and interpret the system-provided Rules of Engagement while planning with the operator.
71%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Passed
No findings from the security scan
Fix and improve this skill with Tessl
tessl review fix ./packages/decepticon/decepticon/skills/standard/soundwave/roe-template/SKILL.mdThe product creates plan/roe.json from engagement setup and manages changes through its own UI and authorization controls. Soundwave may read it, but must never create, edit, replace, or sign it with filesystem tools. If it is absent, ask the operator to complete system setup outside Plan mode before planning target-facing work. If the operator wants a scope or exclusion change, direct them to the system-managed engagement settings; do not silently widen planning documents.
Read the existing RoE before writing other planning documents. Use its exact in-scope and out-of-scope targets, prohibited and permitted actions, authorization source, machine-enforcement rules, and any active time constraints. Do not invent an engagement type or a testing window when the system did not supply one. Ask the operator about choices for Soundwave-owned planning documents only when those choices remain material and unconfirmed.
The RoE constrains the threat profile, CONOPS, deconfliction, contact, data handling, abort, and cleanup documents. A planning draft or complete_engagement_planning call does not authorize a Red run or modify the RoE.
PurpleAILAB/Decepticon
49888ee
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.