BD bomb-dog-sniff
Security-first skill management for OpenClaw - like a bomb-sniffing dog for skills. Sniffs out malicious payloads (crypto stealers, keyloggers, reverse shells) before installation. Quarantine → Scan → Install only the safe ones.
Security-first skill management for OpenClaw - like a bomb-sniffing dog for skills.
As a process D 47/100 · Unfinished process — weak spots: when it triggers, inputs and preconditions, failures and branches
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 purpose itself is risky: wallets, browser password stores, offensive security. Even an honest implementation gives the agent access to things that cost money.
Explain in the description why the access is needed and how it is limited; add tests that show refusals on dangerous requests.
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
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- 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 · 12
✓ No critical or high findings
Medium and low: 12
-
medium Risky intent
intent-wallet-secretspatterns.js:28Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target// Mnemonic phrase extraction
-
medium Dangerous commands
cmd-persistencepatterns.js:227Persistence mechanism (cron / launchd / scheduled task / autorun registry) (string literal in code, not executed)/Library.*LaunchAgents/i,
code literal -
medium Risky intent
intent-wallet-secretsSKILL.md:89Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer targetCrypto wallet private key harvesting detected
-
low Risky intent
intent-wallet-secretspatterns.js:37Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target (quoted — discussed, not commanded)description: 'Crypto wallet private key harvesting detected',
quoted -
low Dangerous commands
cmd-pipe-to-shellpatterns.js:161Downloads and executes remote code from an unrecognised host (pipe to shell) (detector / deny-list definition; string literal in code, not executed; code comment)// HIGH: curl | bash patterns
detectorcode literalcomment -
low Dangerous commands
cmd-pipe-to-shellpatterns.js:171Downloads and executes remote code from an unrecognised host (pipe to shell) (detector / deny-list definition; string literal in code, not executed)description: 'Dangerous curl | bash pattern detected',
detectorcode literal -
low Risky intent
intent-wallet-secretsscripts/scan.js:62Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target (quoted — discussed, not commanded)description: 'Cryptocurrency mnemonic seed phrase handling',
quoted -
low Risky intent
intent-offensive-securitySKILL.md:94Offensive-security / dual-use content (legitimate for authorised testing; review intended use)Reverse shell or remote code execution detected
-
low Risky intent
intent-offensive-securitySKILL.md:205Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| **keylogger** | CRITICAL | Keyboard capture with exfiltration, clipboard theft, password field monitoring |
A further 3 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 9. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-no-whendescription does not say WHEN to use the skill (no "use when") - note
frontmatter-keyunknown frontmatter key "homepage"
Process rating: all ten parameters 47/100
- 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. 1 mutating operations with no state check
- 40Consistency. Frontmatter name (bomb-dog-sniff) differs from the folder (skill-bomb-dog-sniff)
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 100Steps. 84 steps
- 100Execution cost. Instruction body is 3301 tokens
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 16 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)
- -4Absolute local paths (C:\Users, /home/…): not portable
- +1No license
- +2Single-language instructions
- +3Description length 228: enough signal without eating the budget
- +4Structure: 31 headings
- +3Step-by-step instructions: 84 items
- +3Output format is stated explicitly
- +4Has examples (17 code blocks)
- +3All 2 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 73.