Security embedded Dev
v1.0.0Reviews AI/ML model supply chains for security risks including model provenance verification, training data lineage, fine-tuning pipeline integrity, inferenc...
⭐ 0· 99·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The SKILL.md describes a model supply-chain security assessor and the allowed-tools (Read, Grep, Glob) and lack of required env vars/install match that purpose. Minor inconsistency: the registry skill name is "Security embedded Dev" while the SKILL.md identifies itself as "model-supply-chain" — a naming mismatch but not a functional problem.
Instruction Scope
Instructions stay within defensive review scope: they tell the agent to read code/configs, check provenance, and explicitly forbid executing code or following embedded instructions. They limit tool use to Read/Grep/Glob. Note: some recommended checks (cloud IAM, SLSA artifacts, Sigstore proofs) may require out-of-band access or credentials not provided by the skill; the skill correctly does not request those credentials but will produce gaps if the agent lacks access.
Install Mechanism
No install spec and no code files — instruction-only skill. This is the lowest-risk install mechanism and consistent with a review guidance document.
Credentials
The skill requests no environment variables, credentials, or config paths. Given its stated purpose, that is proportionate: it guides checks against artifacts rather than requiring direct credentials. If the user wants deeper checks (e.g., cloud IAM), they will need to provide credentials separately and explicitly.
Persistence & Privilege
always is false and the skill does not request persistent privileges or attempt to modify agent/system configuration. Autonomous invocation is allowed (platform default) but not combined with other concerning privileges.
Assessment
This skill is a defensive, instruction-only checklist for reviewing ML model supply chains and is internally consistent. Before using it: (1) confirm you have authorization to review any target systems and artifacts; (2) be aware the agent will need read access to repo files, CI configs, or model documentation to produce a useful report — it will not and should not request cloud credentials itself; (3) expect gaps where the skill recommends checking cloud IAM or signed provenance unless you explicitly provide access via your normal, secure workflows; (4) note the minor name mismatch (registry name vs SKILL.md name) and verify you intended to install this particular skill; (5) do not grant broad secrets or make the agent execute downloaded model code — the skill already warns against execution.Like a lobster shell, security has layers — review code before you run it.
appsecvk9703pny899ryd1am3014kv5wh8353vacompliancevk9703pny899ryd1am3014kv5wh8353vaidentityvk9703pny899ryd1am3014kv5wh8353valatestvk9703pny899ryd1am3014kv5wh8353vascannervk9703pny899ryd1am3014kv5wh8353vasecurityvk9703pny899ryd1am3014kv5wh8353vasecurityengineervk9703pny899ryd1am3014kv5wh8353vavcisovk9703pny899ryd1am3014kv5wh8353vavulnerabilitiesvk9703pny899ryd1am3014kv5wh8353va
License
MIT-0
Free to use, modify, and redistribute. No attribution required.