Week 45, updated 2 Nov 2026

Agent skills & MCP Agent Skills

Are the Agent Skills safe?

Short answer: No issues found. StackGrade's weekly scan of anthropics/skills found no issues: no critical-severity dependency advisories, no downloading install hooks, no remote-code patterns and no agent instruction file that changes the label. It does list 4 high-severity advisories in runtime dependencies, which don't change the label. 4 phrases in agent instruction files were flagged for human review (unclear). Automated scan of commit ca45fc4 on 2 Nov 2026, not a manual audit.

Updated : the last week its score, rank or safety result changed.

Anthropic's public repository of Agent Skills, reusable instruction packs that agents load on demand.

  • 186,457 stars
  • pushed 28 Oct
  • Apache-2.0 + proprietary
  • Python

github.com/anthropics/skills

53score out of 100

#25of 52 overall#3in Agent skills & MCP▲ 2 vs Week 44

  • Maintenance 30%12
  • Adoption 30%84
  • Docs 20%69
  • Community 20%–
W42W43W44W45

Score over the last 4 weeks: 53, 52, 51, 53.

No issues found

All checks clean in the scan of 2 Nov 2026.

How this score works

Each part is a percentile among all 52 tracked tools, so 50 is average for this list. Community is not measured yet; the score uses the other 80%.

  • Maintenance: Last push 28 Oct 2026. No GitHub releases, no release pace measured. Issue close time: fewer than 5 closed issues, so not used.
  • Adoption: 186,457 stars (70%), 22,028 forks (30%).
  • Docs: README 5,552 characters, 13 headings, 3 code blocks, install section: yes.

Methodology

#3 of 9 in Agent skills & MCP

Ranked by score. The yellow row is this tool.

Agent skills & MCP, Week 45score 0–100, safety
  1. 1Superpowers#13 overall62No dependencies scanned
  2. 2Context7#19 overall57Caution
  3. 3Agent Skills#25 overall53No issues found
  4. 4MCP reference servers#29 overall48No issues found
  5. 5awesome-mcp-servers#30 overall47No dependencies scanned
  6. 6Playwright MCP#32 overall45No issues found
Show the other 3
  1. 7wshobson/agents#38 overall42No issues found
  2. 8GitHub MCP Server#40 overall41No issues found
  3. 9skill-recorder#52 overall22No issues found

Safety scan, 2 Nov 2026

Automated scan of commit ca45fc4, 216 files. Not a manual audit.

No issues found

  • 0Critical dependency advisories4 high, 1 medium, 3 low. OSV, runtime dependencies only.
  • 0Install hooksNo npm install scripts found.
  • 0Remote-code patternsFor example eval(fetch(…)).
  • 212Agent files checked4 phrases flagged for review.
  • 0curl | sh installersIn docs or program code. Information only.
Why this label?

No issues found by the automated scan.

Known vulnerabilities in runtime dependencies (OSV): 0 critical, 4 high, 1 medium, 3 low

Also found (informational, does not change the label):

  • 4 injection-related phrases in instruction files, flagged for human review (often defensive guidance), e.g. skills/claude-api/shared/evals/build-eval.md
  • 4 high-severity known vulnerabilities in runtime dependencies (OSV)
  • Skill scripts contact: api.anthropic.com

Automated scan of commit ca45fc4 on 2 Nov 2026, not a manual audit. 216 files checked. What we check

Repo activity

From the GitHub API, fetched 2 Nov 2026.

  • 0GitHub releasesNo release in the public feed.
  • –Median issue closeFewer than 5 closed issues
  • 186kStars+2,638 since 26 Oct
  • 22kForks1,461 open issues
  • 2025Repo created22 Sept 2025

Questions about Agent Skills

Answers use this week's StackGrade data.

Is Agent Skills safe to use?

No issues found. StackGrade's weekly scan of anthropics/skills found no issues: no critical-severity dependency advisories, no downloading install hooks, no remote-code patterns and no agent instruction file that changes the label. It does list 4 high-severity advisories in runtime dependencies, which don't change the label. 4 phrases in agent instruction files were flagged for human review (unclear). Automated scan of commit ca45fc4 on 2 Nov 2026, not a manual audit.

How does StackGrade score Agent Skills?

Agent Skills scores 53 out of 100 in Week 45: #25 of 52 tracked tools overall and #3 in Agent skills & MCP. The score combines maintenance (12), adoption (84) and docs (69) as percentiles among all 52 tracked tools; community (20% weight) is not measured yet, so the score uses the other 80%. It is not a hands-on test and it is separate from the safety label.

What are the alternatives to Agent Skills?

The highest-scored other tools in Agent skills & MCP this week: Superpowers (62, No dependencies scanned), Context7 (57, Caution), MCP reference servers (48, No issues found), awesome-mcp-servers (47, No dependencies scanned). StackGrade has not tested them hands-on.

What licence does Agent Skills use?

The licence listed for anthropics/skills is Apache-2.0 + proprietary. Read the licence text before commercial use.

Compare and use it

All facts and sources
Type
Agent skill
License
Apache-2.0 + proprietary
Language
Python
Created
22 Sept 2025
Owner
anthropics (Organization)
Forks
22,028

GitHub description: “Public repository for Agent Skills”

Sources: github.com/anthropics/skills, api.github.com/repos/anthropics/skills, raw.githubusercontent.com/anthropics/skills/main/README.md. Fetched 2 Nov 2026.

Know when Agent Skills changes label

One email a week: the new top 10, the biggest score changes and every safety label that changed.

Free. Double opt-in: you confirm by email first, unsubscribe in one click.