Claude Code, Plugin
Claude Ollama Skills
Skills and agents that delegate small tasks to local Ollama models: commit messages, PR descriptions with gated draft-PR creation, shell and code drafts, and log/text/git-history digests.
Compatibility: Not stated.
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 27cdd1d9c69f, unchanged since the check below.
INSTALL_REMOTE_SCRIPT SECRETS_PATH_CREDENTIAL
Why this grade
- INSTALL_REMOTE_SCRIPT Medium
Fetches a script from the network and runs it (a pipe to a shell, or an install hook with a URL).
.github/workflows/e2e.yml, line 18
run: curl -fsSL https://ollama.com/install.sh | sh - SECRETS_PATH_CREDENTIAL Medium
Mentions a credential store path (~/.ssh, id_rsa, .npmrc, .aws/credentials, credentials.json, .netrc).
scripts/ollama_ask.py, line 1433
# NOT here - each can read a secret file (`~/.ssh/id_rsa`, `.env`,
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
Note
requests shell access. The manifest names this permission.
PERMISSION_SHELL_DECLARED
benchmarks/fixtures.py, line 72
subprocess.run(
benchmarks/measure_ab.py, line 65
proc = subprocess.run(argv, cwd=cwd, capture_output=True,
benchmarks/measure_ab.py, line 103
return subprocess.run(["git"] + list(args), cwd=repo, capture_output=True,
Show more
benchmarks/measure_ab.py, line 147
subprocess.run(["git", "-c", "user.name=AB Bench",
benchmarks/measure_ab.py, line 264
health = subprocess.run(
scripts/ollama_ask.py, line 544
exactly -- passing no `env` kwarg to subprocess.run at all. Introduced
scripts/ollama_ask.py, line 553
return subprocess.run(
scripts/ollama_ask.py, line 1579
result = subprocess.run(
Install and update
Serious
install fetches a script from the network.
INSTALL_REMOTE_SCRIPT
.github/workflows/e2e.yml, line 18
run: curl -fsSL https://ollama.com/install.sh | sh
Network
Concern
can reach a host that is not written as a fixed address.
NETWORK_NO_FIXED_HOST
scripts/ollama_ask.py, line 296
with urllib.request.urlopen(host + path, timeout=10) as resp:
scripts/ollama_ask.py, line 328
with urllib.request.urlopen(request, timeout=stall_seconds) as resp:
Secrets
Concern
reads a credential or environment file.
SECRETS_PATH_CREDENTIAL SECRETS_PATH_DOTENV
scripts/ollama_ask.py, line 1414
".ssh", ".aws", ".env", "printenv", "Get-ChildItem Env:",
scripts/ollama_ask.py, line 1433
# NOT here - each can read a secret file (`~/.ssh/id_rsa`, `.env`,
docs/skill-tests.md, line 23
| Baseline (no skill) | Adds `--dry-run` first (good), but final plan still runs **full-repo** `git clean -fdx`, which also deletes ignored files like `.env`. No scope narrowing to
Show more
skills/ollama-docker/DENYLIST.md, line 21
`~/.aws`, `.env`, cloud metadata endpoints.
skills/ollama-shell/DENYLIST.md, line 7
- `git clean -fdx` / `-fd` (silently deletes untracked files - configs, .env, notes)
skills/ollama-shell/DENYLIST.md, line 12
- Reading or sending secrets: credential files (`.ssh`, `.aws`, tokens, browser profiles), `.env` files, or dumping env vars (`printenv`, `Get-ChildItem Env:`)
skills/ollama-shell/DENYLIST.md, line 18
- Docker host-escape the user did not ask for: `--privileged`, `--pid=host`, `--network=host`, `--cap-add=ALL`, `--security-opt seccomp=unconfined`, bind-mounting host root (`-v /:
Agent instructions
Pass
No instruction to ignore earlier instructions or to hide a step.
Obfuscation
Pass
No encoded or minified payload found.
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 27cdd1d9c69f. requests shell access. The manifest names this permission. install fetches a script from the network. can reach a host that is not written as a fixed address. reads a credential or environment file.
Date and source
Date checked:
github.com
Commit graded: 27cdd1d9c69f050e9336819d4c9c9c6902a1fd1b
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.