Claude Code, Skill
Developer Platform Skills
Skills for platform and SDK developer experience — API design, auth and error patterns, integration marketplace strategy, SDK portfolio, app store listings, partner onboarding, and developer observability.
Compatibility: Not stated.
Grade D
Two or more Medium findings, or a High. Only try it in a sandbox.
Not a guarantee. A hygiene check of the source at the commit shown, not a security audit.
INSTALL_REMOTE_SCRIPT PERMISSION_SHELL_UNDECLARED
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).
skills/github-action-offering/references/security-hardening.md, line 40
run: curl -sSL https://get.vendor.dev/install.sh | bash - PERMISSION_SHELL_UNDECLARED Medium
Runs shell commands that the manifest or description does not mention.
evals/_harness/run_skill.py, line 75
proc = subprocess.run(
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
requests shell access. No manifest field names shell or command execution. can delete files.
PERMISSION_SHELL_UNDECLARED PERMISSION_DELETE_FILES
evals/_harness/run_skill.py, line 75
proc = subprocess.run(
evals/_harness/run_skill.py, line 455
shutil.rmtree(root, ignore_errors=True)
evals/_harness/run_skill.py, line 607
"harness_version": subprocess.run(
Install and update
Serious
install fetches a script from the network.
INSTALL_REMOTE_SCRIPT
skills/github-action-offering/references/security-hardening.md, line 40
run: curl -sSL https://get.vendor.dev/install.sh | bash
Network
Concern
can reach a host that is not written as a fixed address.
NETWORK_NO_FIXED_HOST
evals/api-idempotency-retry/trials/eval-5/with_skill/output.md, line 263
const res = await fetch(url, {
evals/api-rate-limit-policy/trials/eval-5/without_skill/output.md, line 172
const res = await fetch(url, init);
evals/api-status-communication/trials/eval-7/without_skill/output.md, line 125
fetch("/uptime.json",{cache:"no-store"}).then(r=>{if(!r.ok)throw 0;return r.json()}).then(d=>{
Show more
evals/oauth2-provider-design/trials/eval-1/with_skill/output.md, line 53
fetch(url, { ...init, headers: { ...init.headers, Authorization: `Bearer ${token}` } });
evals/oauth2-provider-design/trials/eval-1/without_skill/output.md, line 59
fetch(url, { headers: { Authorization: `Bearer ${token}` } })
skills/github-engagement-playbook/scripts/github-utils.gs, line 27
const response = UrlFetchApp.fetch(GITHUB_API + path, {
skills/github-engagement-playbook/scripts/github-utils.gs, line 42
const res = UrlFetchApp.fetch(GITHUB_API + '/graphql', {
skills/github-engagement-playbook/scripts/github-utils.gs, line 64
const res = UrlFetchApp.fetch(`https://github.com${url}`, {
Secrets
Concern
reads a credential or environment file.
SECRETS_PATH_DOTENV
.gitignore, line 33
# Environment
.gitignore, line 34
.env
.gitignore, line 35
.env.local
Show more
evals/api-auth-key-management/trials/eval-8/with_skill/output.md, line 56
- `.env.dist` ships empty placeholders.
evals/api-auth-key-management/trials/eval-8/without_skill/output.md, line 70
| **Create** | Label, scope. The full key shows **once** in a modal with a copy button and a "copy as `.env` line" button. The user must tick "I saved it" to close. Send a "new key
evals/api-auth-key-management/trials/eval-8/without_skill/output.md, line 101
1. **Startup check.** If the pepper, DB URL, or any required secret is missing, the process exits with a clear message: `QUILLHOOK_PEPPER not set. Run 'make dev-env' to generate .e
evals/api-auth-key-management/trials/eval-8/without_skill/output.md, line 102
2. **`make dev-env`** (or `quillhook dev init`) generates random secrets into a git-ignored `.env.local` and ships a `.env.dist`, per the repo's usual convention. First run takes a
evals/api-auth-key-management/trigger-evals.json, line 625
"query": "Migrate our internal service secrets from .env files to Vault",
Agent instructions
Pass
No instruction to ignore earlier instructions or to hide a step.
Obfuscation
Note
the file cap was reached before every file was read.
OBFUSCATION_FILE_CAP_REACHED
evals/bulk-data-sharing-design/trials/eval-7/without_skill/output.md, line 1
# What needs to change before you scale Vantablue's export bucket
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 c72e2a891e65. requests shell access. No manifest field names shell or command execution. can delete files. 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. the file cap was reached before every file was read.
Date and source
Date checked:
github.com
Commit graded: c72e2a891e653dd56334c0fa1c32f046933f5319
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.