Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/hooxuefeng-agent-tool-guardrails
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions: not checked yet.
What's inside
- Skill name
agent-tool-guardrails- Description
- None
- License
- GPL-3.0
- Instructions
- 19 lines in SKILL.md · view on GitHub
Agent Skills spec issues
- invalid YAML frontmatter: Nested mappings are not allowed in compact mappings at line 3, column 14:
- name: missing
- description: missing
Files (1)
- SKILL.md
SKILL.md (first lines)
# Agent Tool Guardrails ## When to Use Use when reviewing how AI can invoke real security tools. ## Workflow 1. Confirm tools are allowlisted and mapped to policy classes. 2. Confirm project scope is checked before every network action. 3. Keep dangerous or unknown actions default-deny. 4. Separate model recommendations from deterministic execution. 5. Preserve Skill Run, Task, Evidence, and Agent Event traces. ## Verification No SKILL.md content can directly become arbitrary shell execution. ## Safety Boundary This skill may only use deterministic capabilities mapped by SnowEdge. The Markdown body is never executed as shell code.
From HooXuefeng/SnowEdge, GPL-3.0.
More from HooXuefeng/SnowEdge
- 0.566score
authorization-differential-review
HooXuefeng
Guides read-only User A/User B/Admin/Anonymous response comparison for object-level and role-level authorization review.
SkillBFreshness: B★ 4Claude Code +9 - 0.566score
browser-observation-workspace
HooXuefeng
Uses a scope-constrained observe-only Chromium session to collect DOM, form, script, Network/XHR/Fetch, console and screenshot evidence without automatic…
SkillBFreshness: B★ 4Claude Code +9 - 0.558score
pentest-coverage-judge
HooXuefeng
Evaluates which safe assessment surfaces have evidence and which selected capabilities remain uncovered.
SkillBFreshness: B★ 4Claude Code +9 - 0.558score
reporting-evidence-pack
HooXuefeng
Structures findings, authorization cases, and evidence into reproducible Markdown/HTML deliverables.
SkillBFreshness: B★ 4Claude Code +9 - 0.558score
request-response-diff
HooXuefeng
Uses the scoped Request Workspace to compare read-only responses while masking identity secrets.
SkillBFreshness: B★ 4Claude Code +9 - 0.558score
safe-reconnaissance
HooXuefeng
Maps authorized hosts and common TCP services using constrained reconnaissance and stores evidence.
SkillBFreshness: B★ 4Claude Code +9 - 0.558score
transport-security-review
HooXuefeng
Reviews HTTPS/TLS posture and evidence-backed security-header hardening observations.
SkillBFreshness: B★ 4Claude Code +9 - 0.558score
web-attack-surface-mapping
HooXuefeng
Collects an HTTP baseline, crawls in-scope HTML with GET only, and builds the web resource graph.
SkillBFreshness: B★ 4Claude Code +9 - 0.558score
evidence-ai-triage
HooXuefeng
Provides the AI model a bounded evidence summary and asks for evidence-backed findings only.
SkillBFreshness: B★ 4Claude Code +9 - 0.558score
javascript-api-mapper
HooXuefeng
Statically reviews discovered JavaScript for API-style routes, fetch/axios calls, and source-map references.
SkillBFreshness: B★ 4Claude Code +9
Signals
- Stars
- 4
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 14, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- GPL-3.0
- Repository
- github.com/HooXuefeng/SnowEdge
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from HooXuefeng
- 0.566score
authorization-differential-review
HooXuefeng
Guides read-only User A/User B/Admin/Anonymous response comparison for object-level and role-level authorization review.
SkillBFreshness: B★ 4Claude Code +9 - 0.566score
browser-observation-workspace
HooXuefeng
Uses a scope-constrained observe-only Chromium session to collect DOM, form, script, Network/XHR/Fetch, console and screenshot evidence without automatic…
SkillBFreshness: B★ 4Claude Code +9 - 0.558score
pentest-coverage-judge
HooXuefeng
Evaluates which safe assessment surfaces have evidence and which selected capabilities remain uncovered.
SkillBFreshness: B★ 4Claude Code +9 - 0.558score
reporting-evidence-pack
HooXuefeng
Structures findings, authorization cases, and evidence into reproducible Markdown/HTML deliverables.
SkillBFreshness: B★ 4Claude Code +9 - 0.558score
request-response-diff
HooXuefeng
Uses the scoped Request Workspace to compare read-only responses while masking identity secrets.
SkillBFreshness: B★ 4Claude Code +9 - 0.558score
safe-reconnaissance
HooXuefeng
Maps authorized hosts and common TCP services using constrained reconnaissance and stores evidence.
SkillBFreshness: B★ 4Claude Code +9
Cite this page
agent-tool-guardrails skill. Agent Mart. https://agentmart.dev/skills/hooxuefeng-agent-tool-guardrails (last updated 2026-10-24)