Security Audit

distributed-llm-pretraining-torchtitan

github.com/davila7/claude-code-templates

AI SkillCommit 458b11867eae

TRUSTED

Scanned 11 days ago

Critical

Immediate action required

High

Priority fixes suggested

Medium

Best practices review

Low

Acknowledged / Tracked

Trust Assessment

distributed-llm-pretraining-torchtitan received a trust score of 83/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.

SkillShield's automated analysis identified 3 findings: 0 critical, 0 high, 2 medium, and 1 low severity. Key findings include Network egress to untrusted endpoints, Covert behavior / concealment directives, Unpinned Git dependency in installation instructions.

The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.

Last analyzed on February 12, 2026 (commit 458b1186). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.

Layer Breakdown

Manifest Analysis

91%

Static Code Analysis

93%

Dependency Graph

100%

LLM Behavioral Safety

100%

Behavioral Risk Signals

Network Access

2 findings

Security Findings3

Severity	Finding	Layer	Location
MEDIUM	Network egress to untrusted endpoints HTTP request to raw IP address Review all outbound network calls. Remove connections to webhook collectors, paste sites, and raw IP addresses. Legitimate API calls should use well-known service domains.	Manifest	cli-tool/components/mcps/devtools/figma-dev-mode.json:4
MEDIUM	Unpinned Git dependency in installation instructions The skill's `SKILL.md` instructions advise users to install `torchao` directly from a Git repository (`git+https://github.com/pytorch/ao.git`) without specifying a commit hash or tag. This bypasses the version constraint (`torchao>=0.5.0`) declared in the skill's manifest and means the installation will always pull the latest `HEAD` of the main branch. This can lead to non-reproducible builds and potential security risks if the upstream repository is compromised or undergoes unexpected changes, as the exact code installed is not fixed. Pin the Git dependency to a specific commit hash or tag (e.g., `git+https://github.com/pytorch/ao.git@<commit_hash>`) to ensure reproducible and stable installations. Alternatively, ensure the PyPI version (`torchao>=0.5.0`) is sufficient and remove the Git installation instruction.	Static	SKILL.md:170
LOW	Covert behavior / concealment directives Multiple zero-width characters (stealth text) Remove hidden instructions, zero-width characters, and bidirectional overrides. Skill instructions should be fully visible and transparent to users.	Manifest	cli-tool/components/mcps/devtools/jfrog.json:4

Scan History

Embed Code

[![SkillShield](https://skillshield.io/api/v1/badge/aec0ef33d3da4dee.svg)](https://skillshield.io/report/aec0ef33d3da4dee)