CC detecting-container-escape-with-falco-rules
Writes and tunes Falco rule syntax for container escape detection - conditions, macros, lists, priorities, and output fields - covering host filesystem mounts, sensitive host path access, kernel module loading, and privileged capability abuse, including how to drive down false positives. Use when authoring or tuning a specific Falco rule for breakout behaviour, or triaging a noisy escape-related Falco alert. Keywords: Falco rule, macro, list, condition, priority, falco_rules.local.yaml, tuning, false positive. Do not use for deploying and operating Falco itself - use detecting-container-runtime-threats-with-falco; for tool-agnostic escape signals use detecting-container-escape-attempts.
Writes and tunes Falco rule syntax for container escape detection - conditions, macros, lists, priorities, and output fields - covering host filesystem…
As a process C 62/100 · Has gaps — weak spots: result and completion, 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 instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".
If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.
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 · 10
✓ No critical or high findings
Medium and low: 10
-
medium Exfiltration
exfil-read-secret-filesreferences/workflows.md:57Reads credential / secret files (documentation of a security skill)kubectl run escape-test-shadow --image=alpine --restart=Never -- cat /etc/shadow
security skill -
low Risky intent
intent-offensive-securityreferences/standards.md:22Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| T1003 | OS Credential Dumping | /etc/shadow access from container |
-
low Risky intent
intent-offensive-securityreferences/standards.md:25Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| T1068 | Exploitation for Privilege Escalation | Kernel exploit indicators |
-
low Risky intent
intent-offensive-securityreferences/standards.md:31Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- Monitor for privilege escalation attempts
-
low Dangerous commands
cmd-privilegeSKILL.md:95Privilege escalation / world-writable permissions (documentation of a security skill)sudo tee /etc/apt/sources.list.d/falcosecurity.list
security skill -
low Dangerous commands
cmd-background-processSKILL.md:101Starts a background / autostarted process (documentation of a security skill)sudo systemctl enable falco
security skill -
low Exfiltration
exfil-webhook-urlSKILL.md:318Webhook / callback URL commonly used for exfiltration (verify the destination) (placeholder value)webhookurl: "https://hooks.slack.com/services/XXXXX"
placeholder
A further 3 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 62/100
- 0Result and completion. Does not say what the result is
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 3 mutating operations with no state check
- 50When it triggers. No condition that starts the skill
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Tools and files. No external tools needed
- 100Steps. 16 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2380 tokens
- low 10 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
- +3Output format is not stated: the model decides each time
- -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
- +4Description says when NOT to use the skill
- +3Description length 695: enough signal without eating the budget
- +4Structure: 22 headings
- +3Step-by-step instructions: 16 items
- +4Has examples (14 code blocks)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 79.