CC hunting-for-persistence-mechanisms-in-windows
Systematically hunts for adversary persistence mechanisms across Windows endpoints, covering registry Run/RunOnce keys, services, startup folders, scheduled tasks, and WMI event subscriptions. Use when performing a broad persistence sweep during incident response or building SIEM detections that cover the full range of Windows persistence techniques (MITRE T1547).
Systematically hunts for adversary persistence mechanisms across Windows endpoints, covering registry Run/RunOnce keys, services, startup folders, scheduled…
As a process C 60/100 · Has gaps — weak spots: when it triggers, failures and branches, running it twice
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 skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.
Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.
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 · 6
✓ No critical or high findings
Medium and low: 6
-
medium Dangerous commands
cmd-persistencereferences/standards.md:63Persistence mechanism (cron / launchd / scheduled task / autorun registry) (documentation of a security skill)HKLM\SOFTWARE\Microsoft\Windows\CurrentVersion\Run
security skill -
medium Dangerous commands
cmd-persistencereferences/standards.md:65Persistence mechanism (cron / launchd / scheduled task / autorun registry) (documentation of a security skill)HKCU\SOFTWARE\Microsoft\Windows\CurrentVersion\Run
security skill -
medium Dangerous commands
cmd-persistencescripts/agent.py:11Persistence mechanism (cron / launchd / scheduled task / autorun registry) (string literal in code, not executed)r"HKLM\SOFTWARE\Microsoft\Windows\CurrentVersion\Run",
code literal -
medium Dangerous commands
cmd-persistencescripts/agent.py:13Persistence mechanism (cron / launchd / scheduled task / autorun registry) (string literal in code, not executed)r"HKCU\SOFTWARE\Microsoft\Windows\CurrentVersion\Run",
code literal -
low Dangerous commands
cmd-persistencereferences/api-reference.md:40Persistence mechanism (cron / launchd / scheduled task / autorun registry) (documentation table row; documentation of a security skill)| `HKLM\...\CurrentVersion\Run` | Auto-start programs |
tablesecurity skill
A further 1 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 8. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "domain" - note
frontmatter-keyunknown frontmatter key "subdomain" - note
frontmatter-keyunknown frontmatter key "d3fend_techniques" - note
frontmatter-keyunknown frontmatter key "nist_csf" - note
frontmatter-keyunknown frontmatter key "mitre_attack"
Process rating: all ten parameters 60/100
- 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. 7 mutating operations with no state check
- 60Tools and files. Uses tools (bash) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Steps. 22 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 986 tokens
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)
- -43 reference files, but SKILL.md never points to them: the model will not open them
- -32 of 2 scripts are never mentioned in SKILL.md
- +2Single-language instructions
- +3Description length 366: enough signal without eating the budget
- +4Structure: 8 headings
- +3Step-by-step instructions: 22 items
- +3Output format is stated explicitly
- +4Has examples (1 code blocks)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 78.