DC gkn-phantom
GKN-Phantom — production-grade automated penetration testing and security validation skill. v5.3 completes the Quick Combat impact chain: quick detect (nuclei + built-in probes) → content-aware severity escalation (v5.2) → deep-dive adapters proving impact from exposed content (Actuator/GraphQL/Swagger, v5.3 C1) → parameter discovery with bounded error-based SQLi/SSTI probing (v5.3 C2) → auto PoC generation → auto exploit generation → bundled combat report. Retains v5 capabilities including interactive browser agent (Playwright), HTML/PDF visualization reports, YAML rule configuration with hot-reload, and full pytest test suite (30+ modules, 38 vulnerability types, 25+ WAF fingerprints, 20+ attack chain patterns). Trigger phrases include penetration test, pentest, security audit, vulnerability scan, security validation, and their Chinese equivalents.
GKN-Phantom — production-grade automated penetration testing and security validation skill.
As a process C 56/100 · Has gaps — weak spots: inputs and preconditions, execution cost
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The files contain someone else's key or token. If it is live, your agent will call third-party services under a stranger's identity; if it was revoked, the skill's scripts simply fail. Such a key often arrives with the author's whole workspace, personal data included.
The key is visible to everyone who downloaded the skill and has likely been copied by catalog-scanning bots already. Revoke it now, check bills and access logs, then reissue.
How to improve
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- 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 · 18
✓ No critical or high findings
Medium and low: 18
-
medium Secrets in code
secret-private-keyscripts/passive_recon.py:1060Private key material (key header without key body; quoted — discussed, not commanded)("Private keys", '"{domain}" -----BEGIN PRIVATE KEY----- …header onlyquoted -
low Risky intent
intent-offensive-securityREADME.md:3Offensive-security / dual-use content (legitimate for authorised testing; review intended use)> **Production-grade automated penetration testing skill for the OpenClaw AI Agent Framework.**
-
low Dangerous commands
cmd-shell-rcREADME.md:243Writes to a shell startup file (documentation of a security skill)echo 'export PATH=$PATH:$HOME/go/bin' >> ~/.zshrc && source ~/.zshrc
security skill -
low Risky intent
intent-offensive-securityREADME.md:259Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (quoted — discussed, not commanded)user: "Run a penetration test on dev-internal.test"
quoted -
low Risky intent
intent-offensive-securityreferences/data_schemas.md:1Offensive-security / dual-use content (legitimate for authorised testing; review intended use)# Data Schemas — GKN-Phantom Penetration Testing Skill
-
low Risky intent
intent-offensive-securityreferences/payload_playbook.md:11Offensive-security / dual-use content (legitimate for authorised testing; review intended use)3. L4 actions (write methods, internal SSRF sinks, privilege escalation)
-
low Risky intent
intent-offensive-securityreferences/payload_playbook.md:304Offensive-security / dual-use content (legitimate for authorised testing; review intended use)### Privilege Escalation (`priv_esc`)
-
low Risky intent
intent-offensive-securityreferences/safety_policy.md:1Offensive-security / dual-use content (legitimate for authorised testing; review intended use)# Safety Policy — GKN-Phantom Penetration Testing Skill
-
low Risky intent
intent-offensive-securityreferences/safety_policy.md:64Offensive-security / dual-use content (legitimate for authorised testing; review intended use)that mutates server state), privilege escalation attempts, SSRF to internal
-
low Secrets in code
secret-high-entropy-tokenscripts/cloud_security.py:92High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)"headers": {"X-aw…nds": "21600"},quoted -
low Dangerous commands
cmd-privilegescripts/nuclei_runner.py:379Privilege escalation / world-writable permissions (string literal in code, not executed)"unzip nuclei.zip && chmod +x nuclei && sudo mv nuclei /usr/local/bin/",
code literal -
low Dangerous commands
cmd-eval-dynamicscripts/waf_evasion.py:347Dynamic code execution from decoded/untrusted input (detector / deny-list definition)"description": "Wrap in eval(atob(<base64>)) to hide payload",
detector -
low Dangerous commands
cmd-eval-dynamicscripts/waf_evasion.py:348Dynamic code execution from decoded/untrusted input (quoted — discussed, not commanded)"generator": lambda p: f"eval(atob('{_bas…rap(p)}'))",quoted -
low Risky intent
intent-offensive-securitySKILL.md:4Offensive-security / dual-use content (legitimate for authorised testing; review intended use)description: GKN-Phantom — production-grade automated penetration testing and security validation skill. v5.3 completes the Quick Combat impact chain: quick detect (nuclei + built-in probes) → content
-
low Risky intent
intent-offensive-securitySKILL.md:7Offensive-security / dual-use content (legitimate for authorised testing; review intended use)# GKN-Phantom — Penetration Testing Skill
-
low Risky intent
intent-offensive-securitySKILL.md:11Offensive-security / dual-use content (legitimate for authorised testing; review intended use)Execute automated, authorized penetration testing and security validation in
-
low Risky intent
intent-offensive-securitySKILL.md:84Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- A user requests a penetration test, security audit, or vulnerability scan of
-
low Risky intent
intent-offensive-securitySKILL.md:479Offensive-security / dual-use content (legitimate for authorised testing; review intended use)metadata, SQLi→credential dumping, IDOR→mass PII exposure, SSTI→RCE,
Files scanned: 54. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
frontmatter-yamlSKILL.md: the frontmatter is not valid YAML (YAML parse error: Nested mappings are not allowed in compact mappings at line 3, column 14: description: GKN-Phantom — production-grade automated penetration testing and s… ^ ); fields were read line by line. The usual cause is a colon inside an unquoted value - warning
body-longSKILL.md body ≈ 12159 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 56/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 40Execution cost. Instruction body is 12159 tokens: crowds the task out of the window
- 50When it triggers. No condition that starts the skill
- 50Failures and branches. 0 branches, has a failure section
- 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
- 60Steps. 150 steps, 5 vague phrases
- 60Result and completion. Output format stated, no completion criterion
- 100Consistency. Name and required fields are in place
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 14 top-level sections: this looks like several domains in one skill
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
- +3Description length 862: 120–800 characters recommended
- +1No license
- +2Single-language instructions
- +4Description says when NOT to use the skill
- +4Structure: 26 headings
- +3Step-by-step instructions: 150 items
- +3Output format is stated explicitly
- +4Has examples (3 code blocks)
- +4Reference files are cited in the instructions (6 of 6)
- +3All 34 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 75.