Claude Code, Plugin
Amd Skills
AMD's verified Agent Skills in one plugin: route image/audio through local AI on Ryzen AI, serve LLMs on AMD Instinct GPUs with vLLM, optimize inference throughput with Hyperloom, and analyze GPU kernel and PyTorch trace performance.
Compatibility: Portable across standard formats.
Grade D
Two or more Medium findings, or a High finding. Only try it in a sandbox.
Not a guarantee. A hygiene check of the source at the commit shown, not a security audit.
Rescanned at commit c729955ac862.
INSTALL_REMOTE_SCRIPT
Why this grade
- INSTALL_REMOTE_SCRIPT High
Fetches a script from the network and runs it (a pipe to a shell, or an install hook with a URL).
rocm-doctor/SKILL.md, line 88
curl -fsSL https://raw.githubusercontent.com/ROCm/rocm-cli/main/install.sh | sh
Any High, or two or more Medium findings, give D. One Medium gives C. Low findings or a security Note give B. Each code names a finding behind the letter.
If you think this grade is wrong, email [email protected] with the entry link and your evidence. We aim to acknowledge within two working days and decide within seven. If we confirm an error, we aim to correct it within 24 hours of confirming it. These are aims, not guarantees.
Grades are a hygiene check of the source at the commit shown, on the date checked, not a security audit or a guarantee. The free directory grades are automated. Before release, a PlugVet engineer reviews every paid report and any grade we hold back, including all F grades, against the source code and the evidence we quote.
What was checked
Plain notes from the hygiene checks.
Permissions
Concern
can delete files.
PERMISSION_DELETE_FILES
lemonade-router-builder/scripts/validate.py, line 119
result = subprocess.run([node, "-e", script], capture_output=True, text=True, timeout=5)
local-ai-app-integration/reference.md, line 298
def start_lemond(retries: int = 3) -> tuple[subprocess.Popen, str, int]:
local-ai-app-integration/reference.md, line 304
proc = subprocess.Popen(
Show more
local-ai-use/scripts/setup_local_ai.py, line 190
result = subprocess.run(cmd, capture_output=True, text=True, timeout=20)
local-ai-use/scripts/setup_local_ai.py, line 303
result = subprocess.run(cmd, shell=shell) # noqa: S602,S603
serving-llms-on-epyc/scripts/check_model.py, line 177
r = subprocess.run(["python", "-c", snippet], stdout=subprocess.PIPE,
serving-llms-on-epyc/scripts/cpu_tune.py, line 53
r = subprocess.run(cmd, shell=True, stdout=subprocess.PIPE,
serving-llms-on-epyc/scripts/detect.py, line 43
r = subprocess.run(cmd, shell=True, stdout=subprocess.PIPE,
serving-llms-on-instinct/scripts/sync_recipes.py, line 64
shutil.rmtree(tmpdir, ignore_errors=True)
serving-llms-on-instinct/scripts/sync_recipes.py, line 150
shutil.rmtree(repo_dir, ignore_errors=True)
tracelens-analysis-orchestrator/evals/skills/eval-post-processing/run_post_processing.py, line 862
shutil.rmtree(REPRODUCERS_DIR)
tracelens-analysis-orchestrator/evals/skills/eval-post-processing/run_post_processing.py, line 951
shutil.rmtree(LATEST_DIR)
Install and update
Serious
install fetches a script from the network. The docs document an installer that pipes a download into a shell.
INSTALL_REMOTE_SCRIPT INSTALL_DOCUMENTED_INSTALLER
rocm-doctor/SKILL.md, line 88
curl -fsSL https://raw.githubusercontent.com/ROCm/rocm-cli/main/install.sh | sh
tracelens-analysis-orchestrator/evals/eval_scripts/run_pi_analysis_in_docker.sh, line 511
curl -fsSL "https://deb.nodesource.com/setup_${NODE_MIN_MAJOR}.x" | bash -
tracelens-analysis-orchestrator/evals/eval_scripts/run_pi_analysis_in_docker.sh, line 514
curl -fsSL "https://rpm.nodesource.com/setup_${NODE_MIN_MAJOR}.x" | bash -
Show more
tracelens-analysis-orchestrator/evals/eval_scripts/run_pi_analysis_in_docker.sh, line 517
curl -fsSL "https://rpm.nodesource.com/setup_${NODE_MIN_MAJOR}.x" | bash -
tracelens-analysis-orchestrator/evals/README.md, line 50
curl https://cursor.com/install -fsS | bash
Network
Concern
can reach a host that is not written as a fixed address. can reach api.github.com. The description does not name that host.
NETWORK_NO_FIXED_HOST NETWORK_UNNAMED_HOST
local-ai-app-integration/SKILL.md, line 169
URL=$(curl -s https://api.github.com/repos/lemonade-sdk/lemonade/releases/latest \
local-ai-app-integration/SKILL.md, line 171
curl -L "$URL" | tar -xz --strip-components=1 -C vendor/lemonade
local-ai-app-integration/reference.md, line 324
with urllib.request.urlopen(req, timeout=1) as r:
Show more
local-ai-app-integration/reference.md, line 375
const r = await fetch(url, { headers });
local-ai-use/scripts/setup_local_ai.py, line 159
with urllib.request.urlopen(req, timeout=timeout_s) as r: # noqa: S310
local-ai-use/scripts/setup_local_ai.py, line 374
with urllib.request.urlopen(req, timeout=15.0) as r: # noqa: S310
local-ai-use/templates/local-ai-rule.md, line 39
curl -sX POST {{LEMONADE_BASE_URL}}/images/generations \
local-ai-use/templates/local-ai-rule.md, line 58
curl -sX POST {{LEMONADE_BASE_URL}}/audio/speech \
Secrets
Concern
reads a credential or environment file.
SECRETS_ENV_READ SECRETS_PATH_DOTENV
local-ai-app-integration/evals/files/apikey-guard/main.py, line 5
api_key = os.environ.get("OPENAI_API_KEY", "")
serving-llms-on-epyc/scripts/check_model.py, line 199
token = os.environ.get("HF_TOKEN", "")
serving-llms-on-epyc/scripts/estimate_memory.py, line 114
token = os.environ.get("HF_TOKEN", "")
Show more
serving-llms-on-epyc/scripts/validate.py, line 266
if not os.environ.get("HF_TOKEN"):
serving-llms-on-instinct/scripts/estimate_vram.py, line 175
token = os.environ.get("HF_TOKEN", "")
hyperloom-workload-optimizer/SKILL.md, line 61
updates `.env` in it. Do not switch to another directory on your own, and do not
hyperloom-workload-optimizer/SKILL.md, line 62
install into an existing project unless the user accepts the `.env` change.
hyperloom-workload-optimizer/SKILL.md, line 85
framework install. It writes `.env` and stops before any optimization. Run it once
hyperloom-workload-optimizer/SKILL.md, line 110
To resume a stopped session, follow the optimizer skill at `.env`
Agent instructions
Pass
No instruction to ignore earlier instructions or to hide a step.
Obfuscation
Note
a file was over the read cap and was not fully read.
OBFUSCATION_FILE_OVER_READ_CAP
serving-llms-on-instinct/data/recipes_cache.json, line 1
{
Licence
Pass
A licence file is present (LICENSE).
Maintainer activity
Pass
The latest commit is within the last 365 days.
Note
Hygiene check of commit c729955ac862. can delete files. install fetches a script from the network. can reach a host that is not written as a fixed address. can reach api.github.com. The description does not name that host. reads a credential or environment file. a file was over the read cap and was not fully read.
Date and source
Date checked:
github.com
Commit graded: c729955ac862b010f269be9ece99540a2b6cee90
This link goes to the original host. PlugVet does not host or check what you download from there, and installing it is your own decision and risk.