Claude Code, Skill
skillassay
A static analyzer for AI coding-agent context, and an Agent Skill your agent can run itself. It measures the always-on context-window cost of CLAUDE.md, AGENTS.md, GEMINI.md and Agent Skills, then reports skills that will not load, redundant instructions, stale path ref
Compatibility: Not stated.
Grade B
Minor issues only, no Medium or High. Worth a quick look at the findings below before installing.
Not a guarantee. A hygiene check of the source at the commit shown, not a security audit.
Rescanned at commit 3877296fd29b, unchanged since the check below.
NETWORK_NO_FIXED_HOST NETWORK_UNNAMED_HOST SECRETS_PATH_DOTENV PERMISSION_SHELL_DECLARED SECRETS_ENV_READ
Why this grade
- NETWORK_NO_FIXED_HOST Low
Makes a network call whose host is not written as a fixed address.
test/fixtures/anthropic-official/skills/claude-api/curl/examples.md, line 16
curl https://api.anthropic.com/v1/messages \ - NETWORK_UNNAMED_HOST Low
Calls a fixed host that the description does not name.
test/fixtures/anthropic-official/skills/claude-api/curl/examples.md, line 16
curl https://api.anthropic.com/v1/messages \ - SECRETS_PATH_DOTENV Low
Mentions a .env file.
test/fixtures/anthropic-official/skills/claude-api/shared/anthropic-cli.md, line 65
# .env format - sets ANTHROPIC_AUTH_TOKEN (and ANTHROPIC_BASE_URL if the profile has one). - PERMISSION_SHELL_DECLARED Note
Runs shell commands, and the manifest or description says so.
test/fixtures/anthropic-official/skills/claude-api/shared/token-counting.md, line 51
before = subprocess.check_output(["git", "show", "HEAD:CLAUDE.md"], text=True) - SECRETS_ENV_READ Note
Reads an environment variable whose name contains KEY, SECRET, TOKEN, PASSWORD or CREDENTIAL.
test/fixtures/anthropic-official/skills/claude-api/shared/managed-agents-environments.md, line 169
authorization_token: process.env.GITHUB_TOKEN, // repo clone token (≠ MCP auth)
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.
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
test/fixtures/anthropic-official/skills/claude-api/shared/token-counting.md, line 51
before = subprocess.check_output(["git", "show", "HEAD:CLAUDE.md"], text=True)
Install and update
Pass
No install-time script found, and dependencies are pinned or absent.
Network
Concern
can reach api.anthropic.com. The description does not name that host. can reach a host that is not written as a fixed address.
NETWORK_NO_FIXED_HOST NETWORK_UNNAMED_HOST
test/fixtures/anthropic-official/skills/claude-api/curl/examples.md, line 16
curl https://api.anthropic.com/v1/messages \
test/fixtures/anthropic-official/skills/claude-api/curl/examples.md, line 37
response=$(curl -s https://api.anthropic.com/v1/messages \
test/fixtures/anthropic-official/skills/claude-api/curl/examples.md, line 63
curl https://api.anthropic.com/v1/messages \
Show more
test/fixtures/anthropic-official/skills/claude-api/curl/examples.md, line 102
curl https://api.anthropic.com/v1/messages \
test/fixtures/anthropic-official/skills/claude-api/curl/examples.md, line 127
curl https://api.anthropic.com/v1/messages \
test/fixtures/anthropic-official/skills/claude-api/curl/examples.md, line 165
curl https://api.anthropic.com/v1/messages \
test/fixtures/anthropic-official/skills/claude-api/curl/examples.md, line 191
curl https://api.anthropic.com/v1/messages \
test/fixtures/anthropic-official/skills/claude-api/curl/examples.md, line 216
response=$(curl -s https://api.anthropic.com/v1/messages \
test/fixtures/anthropic-official/skills/claude-api/shared/anthropic-cli.md, line 24
curl -fsSL "https://github.com/anthropics/ant
Secrets
Concern
reads a credential or environment file.
SECRETS_ENV_READ SECRETS_PATH_DOTENV
scripts/anti-simulation.mjs, line 34
pattern: /\b(?:demoMode|DEMO_MODE|process\.env\.DEMO|IS_DEMO|MOCK_MODE|process\.env\.MOCK)\b/,
test/fixtures/anthropic-official/skills/claude-api/shared/anthropic-cli.md, line 65
# .env format - sets ANTHROPIC_AUTH_TOKEN (and ANTHROPIC_BASE_URL if the profile has one).
test/fixtures/anthropic-official/skills/claude-api/shared/managed-agents-environments.md, line 169
authorization_token: process.env.GITHUB_TOKEN, // repo clone token (≠ MCP auth)
Show more
test/fixtures/anthropic-official/skills/claude-api/shared/managed-agents-self-hosted-sandboxes.md, line 60
const environmentKey = process.env.ANTHROPIC_ENVIRONMENT_KEY!;
test/fixtures/anthropic-official/skills/claude-api/shared/managed-agents-tools.md, line 92
{ "type": "user.tool_confirmation", "tool_use_id": "sevt_def456", "result": "deny", "message": "Read .env.example instead" }
test/fixtures/anthropic-official/skills/claude-api/typescript/managed-agents/README.md, line 109
authorization_token: process.env.GITHUB_TOKEN,
test/fixtures/anthropic-official/skills/mcp-builder/reference/node_mcp_server.md, line 707
if (!process.env.EXAMPLE_API_KEY) {
test/fixtures/anthropic-official/skills/mcp-builder/reference/node_mcp_server.md, line 719
if (!process.env.EXAMPLE_API_KEY) {
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 3877296fd29b. requests shell access. The manifest names this permission. can reach api.anthropic.com. The description does not name that host. 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: 3877296fd29b4b22f08196aab291498706dac782
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.