BD ctf-writeup-generator
Automatically generate professional CTF writeups from solving sessions with flag detection, challenge categorization, and proper markdown formatting
Automatically generate professional CTF writeups from solving sessions with flag detection, challenge categorization, and proper markdown formatting
As a process D 47/100 · Unfinished process — weak spots: result and completion, inputs and preconditions, failures and branches
AnalyzerSoftware developmentWriting and documentsAI and agentstype and topics are labelled automatically from the skill text
How to improve
For the model run — optional
- 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 · 16
✓ No critical or high findings
Medium and low: 16
-
low Risky intent
intent-offensive-securityexample_writeup.md:12Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (test fixture / example file)This Sherlock challenge involved analyzing Windows Event Logs and network traffic to identify unauthorized remote access to a corporate workstation. The investigation revealed a compromised RDP sessio
fixture -
low Risky intent
intent-offensive-securityexample_writeup.md:85Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (test fixture / example file)2. Previous credential harvesting
fixture -
low Risky intent
intent-offensive-securityexample_writeup.md:123Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (test fixture / example file)# Credential dumping attempt
fixture -
low Dangerous commands
cmd-persistenceexample_writeup.md:154Persistence mechanism (cron / launchd / scheduled task / autorun registry) (test fixture / example file; documentation of a security skill)HKCU\Software\Microsoft\Windows\CurrentVersion\Run\WindowsUpdate
fixturesecurity skill -
low Risky intent
intent-offensive-securityexample_writeup.md:196Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (test fixture / example file; quoted — discussed, not commanded)The flag was embedded in the attacker's PowerShell script downloaded from the C2 server. Extracted the script from network traffic:
fixturequoted -
low Secrets in code
secret-high-entropy-tokenexample_writeup.md:208High-entropy token-like string (may be an id, hash or a credential) (test fixture / example file)# Mission accomplished - HTB{RDP_…m3d}fixture -
low Risky intent
intent-offensive-securityexample_writeup.md:217Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (test fixture / example file)| 10:25:03 | Privilege escalation to admin | Security.evtx (4672) |
fixture -
low Dangerous commands
cmd-persistenceexample_writeup.md:246Persistence mechanism (cron / launchd / scheduled task / autorun registry) (test fixture / example file; documentation of a security skill)HKCU\Software\Microsoft\Windows\CurrentVersion\Run\WindowsUpdate
fixturesecurity skill -
low Secrets in code
secret-high-entropy-tokenexample_writeup.md:259High-entropy token-like string (may be an id, hash or a credential) (test fixture / example file)HTB{RDP_…m3d}fixture -
low Risky intent
intent-offensive-securitySKILL.md:8Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- pentesting
-
low Risky intent
intent-offensive-securitySKILL.md:52Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- **PWN**: Exploitation, shellcode, privilege escalation
-
low Risky intent
intent-offensive-securitySKILL.md:187Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- Document privilege escalation chains
A further 4 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 4. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "homepage"
Process rating: all ten parameters 47/100
- 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
- 30Running it twice. 4 mutating operations with no state check
- 60Tools and files. Uses tools (bash) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 85Steps. 103 steps, 1 vague phrases
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2351 tokens
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 15 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
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +3Description length 148: enough signal without eating the budget
- +4Structure: 26 headings
- +3Step-by-step instructions: 103 items
- +4Has examples (9 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 83.