CD security-threat-model
Systematically identifies vulnerabilities, threats, and mitigations for systems handling sensitive data using STRIDE methodology, trust boundary mapping, and defense-in-depth principles. Use when designing or reviewing systems with PII/PHI/financial/auth data, building security-sensitive features (auth, payments, file uploads, APIs), preparing for audits or compliance (PCI, HIPAA, SOC 2), investigating incidents, or integrating third-party services. Use when user mentions threat model, STRIDE, trust boundaries, attack surface, or security review.
Systematically identifies vulnerabilities, threats, and mitigations for systems handling sensitive data using STRIDE methodology, trust boundary mapping, and…
As a process D 43/100 · Unfinished process — weak spots: result and completion, when it triggers, inputs and preconditions
How to improve
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 9
✓ No critical or high findings
Medium and low: 9
-
low Risky intent
intent-offensive-securityresources/methodology.md:32Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)│ ├── SQL injection → shell → lateral movement
detector -
low Risky intent
intent-offensive-securityresources/methodology.md:212Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| Admin panel | Manage system | Privilege escalation, mass data modification | Strong auth (MFA), audit logs, confirmation dialogs, IP whitelisting |
-
low Risky intent
intent-offensive-securityresources/methodology.md:303Offensive-security / dual-use content (legitimate for authorised testing; review intended use)**Tactics** (why): Initial Access, Execution, Persistence, Privilege Escalation, Defense Evasion, Credential Access, Discovery, Lateral Movement, Collection, Exfiltration, Impact
-
low Risky intent
intent-offensive-securityresources/methodology.md:317Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- **Red teaming**: Use ATT&CK techniques to simulate attacks during testing
-
low Risky intent
intent-offensive-securityresources/methodology.md:368Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- **Privilege escalation via misconfigured IAM**: Attacker modifies own permissions
-
low Risky intent
intent-offensive-securityresources/template.md:220Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- **Monitoring**: Alert on privilege escalation attempts, admin action audit log
-
low Risky intent
intent-offensive-securityresources/template.md:324Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| Privilege escalation attempts | Application audit log | Non-admin accessing admin endpoints | Block user, security review | Engineering | ✓ Done |
-
low Risky intent
intent-offensive-securitySKILL.md:78Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)- **Examples**: SQL injection to admin, IDOR to other user data, path traversal, privilege escalation bugs
detector -
low Risky intent
intent-offensive-securitySKILL.md:113Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- **Critical threats**: Service impersonation (Spoofing), lateral movement (Elevation of Privilege), message tampering (Tampering), service enumeration (Information Disclosure)
Files scanned: 4. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 43/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 2 mutating operations with no state check
- 60Tools and files. Uses tools (web) that frontmatter does not declare
- 100Steps. 66 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2332 tokens
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- -215 emoji in the instructions: noise for the model
- +1No license
- +2Single-language instructions
- +3Description length 552: enough signal without eating the budget
- +4Structure: 8 headings
- +3Step-by-step instructions: 66 items
- +4Has examples (1 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 82.