scope-grill
Skill37 starsChanged 5 months agoMITView on GitHub
Interviews the user about a pentest or audit engagement before any testing begins, capturing target, scope, rules of engagement, auth, and deliverables into a structured brief. Trigger when the user says "start a pentest", "begin an engagement", or "test this target", describes a target without mentioning scope/authorization/deliverables, or wants to structure an engagement before diving in.
In this repository38 other files it gives its agents.
- ad-breachSkill
- ai-llm-hunterSkill
- attack-surfaceSkill
- bugbounty-reporterSkill
- cavemanSkill
- check-exploitSkill
- clickjacking-hunterSkill
- compliance-gap-analyzerSkill
- control-lookupSkill
- ctf-writeupSkill
- cvss-scorerSkill
- deadangleSkill
- droid-reconSkill
- economist-attackSkill
- engagement-handoffSkill
- finding-writerSkill
- find-skillsSkill
- graphql-raiderSkill
- hexstrike-forgeSkill
- high-severity-hunterSkill
- hpp-hunterSkill
- idor-hunterSkill
- js-analyzerSkill
- jwt-crackerSkill
- less-aggressive-attackSkill
- less-noise-attackSkill
- nuclei-template-writerSkill
- pentest-reportSkill
- redirect-forgeSkill
- redmindSkill
- remediation-plannerSkill
- risk-assessorSkill
- skill-benchmarkSkill
- ssrf-hunterSkill
- ssti-hunterSkill
- vuln-diagnoseSkill
- xss-hunterSkill
- xxe-phantomSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 13
Before any testing begins, interview the user to build a complete engagement brief.
- Distinctive: line 15
Ask one question at a time.
- Distinctive: line 19
Ask in this order. Skip a question only if the answer is already obvious from context.
- Very specific: line 34
After all questions are answered, produce a filled Engagement Brief using the template in references/engagement-brief.md.
- Distinctive: line 40
Do not begin testing until authorization is confirmed
- Distinctive: line 41
If the user cannot confirm authorization, stop and explain the risk
- Distinctive: line 42
If scope is vague, flag it — an unclear scope is itself a risk
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Scope Grill
Before any testing begins, interview the user to build a complete engagement brief. An untested assumption about scope or authorization invalidates findings and creates legal risk.
Ask one question at a time. Wait for the answer before moving on. Provide a recommended answer or example for each question.
Question Sequence
Ask in this order. Skip a question only if the answer is already obvious from context.
- Target — What is the exact target? (domain, IP range, application name, cloud account)
- Authorization — Do you have written authorization to test this target? (required — do not proceed if no)
- Scope — What is explicitly in scope? What is explicitly out of scope?
- Test type — Black-box (no credentials, no source), grey-box (credentials only), or white-box (full access + source)?
- Accounts — Do you have test accounts? What roles are available? (admin, user, guest, API key?)
- Rules of engagement — Any restrictions? (no DoS, no phishing, no persistence, specific time windows?)
- Known stack — Any known technologies? (framework, language, cloud provider, WAF, CDN?)
- Prior work — Previous pentest reports, known issues, or areas to re-verify?
- Deliverables — What's expected at the end? (findings list, full report, executive summary?) Who is the audience?
- Deadline — When are findings due?
Output
After all questions are answered, produce a filled Engagement Brief using the template in references/engagement-brief.md.
Save it as ENGAGEMENT.md in the current directory unless the user specifies otherwise.
Rules
- Do not begin testing until authorization is confirmed
- If the user cannot confirm authorization, stop and explain the risk
- If scope is vague, flag it — an unclear scope is itself a risk
- One question at a time — do not dump the full list at once
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

