skill-template
One to three sentences describing what this skill does and the conditions under which an agent should activate it. Start with a verb, name the tools and artifacts involved, and end with the trigger ph
Files3
View on GitHubSKILL.mdtemplates/skill-template/SKILL.md
GitHubSkill Title
Replace every bracketed placeholder. Delete this quote block and the
Safety & Handlingsection if the skill does not touch live malware samples. Keep section order.
When to Use
- [Concrete trigger: a situation where this skill is the right tool]
- [Another trigger]
- [Another trigger]
Do not use for [explicit out-of-scope cases]; instead reach for [the correct skill].
Prerequisites
- [Tool + minimum version, with install/download pointer]
- [Lab/environment requirement]
- [Prior knowledge or input artifact required]
Safety & Handling
(Sample-handling skills only — delete otherwise.)
- Work only inside an isolated analysis VM with no path to production or the internet unless deliberately routed through a controlled gateway.
- Store samples encrypted or in password-protected archives (password
infected); never leave live samples loose on disk. - Defang indicators (
hxxp://,1.2.3[.]4) in any output that leaves the lab. - Never execute a sample on the host or detonate outside the sandbox.
Workflow
Step 1: [Action]
[What to do and why.]
# Real, runnable command — not pseudocode
tool --do-the-thing sample.bin
Step 2: [Action]
[Explanation.]
# Original example code
Step 3: [Action]
[Explanation.]
Validation
- [How to confirm the analysis/hunt produced a correct, trustworthy result]
- [A cross-check or second method to corroborate findings]
Pitfalls
- [Common mistake and how to avoid it]
- [Evasion trick or false-positive trap to watch for]
References
- See
references/api-reference.mdfor tooling details and the full list of sources this skill was synthesized from. - [Key external source — title and URL]