security-scan

A security checklist for agent configuration files, including skills, hooks, MCP connections, and settings. It focuses on secrets, permissions, privacy-protected file reading, and repository checks.

In plain words
What is it for?
Use it before enabling automation, after changing agent configuration, or before release to enforce privacy checks, inspect configuration safely, and run the repository’s security verifier.
Why use it?
It reduces the risk of exposing API keys, tokens, cookies, or unsafe automation settings when onboarding a project or preparing configuration changes for release.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/rexleimo/aios/security-scan
Any agent
npx skills add rexleimo/aios --skill security-scan
Clone the repo
git clone --depth 1 https://github.com/rexleimo/aios

Made for: Claude Code, Codex.

Per session 31 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 537 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00031 $0.00537
Opus 5 $0.00015 $0.00269
Sonnet 5 $0.00006 $0.00107
Haiku 4.5 $0.00003 $0.00054

Measured 3d ago against content hash 84f756aaf1ce, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

security-scan scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Downloads and executes remote codemediumSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

- Avoid `curl | bash`, silent error suppression, and untrusted string interpolation.

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- Avoid `curl | bash`, silent error suppression, and untrusted string interpolation.
skill-sources/security-scan/SKILL.md · 58 lines

What it actually says

Security Scan

Trigger

Use this skill when:

  • Onboarding a new repo with agent automation enabled
  • Editing skills, hooks, MCP config, or client settings
  • Before publishing/releasing changes that affect other projects

Mandatory Privacy Rule (No Bypass)

  • For config-like files that may contain keys/tokens/passwords, do not read raw file content directly.
  • Always read through Privacy Guard first:
    • aios privacy read --file <path> (preferred)
    • or node scripts/privacy-guard.mjs read --file <path>
  • If Privacy Guard reports guard-disabled for sensitive files, enable it first:
    • aios privacy enable

Quick Path

  • Ensure strict mode is on: aios privacy enforce-on
  • Run the repo verifier: aios doctor (preferred when shell integration is installed).
  • Or run: scripts/doctor-security-config.sh / scripts/doctor-security-config.ps1.

What To Check (Minimum)

  1. Secrets

    • No API keys/tokens/cookies committed to git.
    • Prefer env vars + local-only config files ignored by git.
    • Use Privacy Guard read output when inspection is required.
    • If scanning finds secrets, rotate them, then remove from history if needed.
  2. Tool/permission scope

    • Avoid wildcard allowlists (when the client supports allow/deny config).
    • Prefer least-privilege tool access and explicit denies for destructive actions.
  3. Hooks safety

    • Avoid curl | bash, silent error suppression, and untrusted string interpolation.
    • Treat any network egress from hooks as high risk; make it opt-in.
  4. MCP / supply chain

    • Prefer pinned versions and explicit installs over ad-hoc npx execution.
    • Review MCP servers for network + filesystem access and log behavior.

Output Discipline

  • Never paste secret values into chat logs or commits.
  • Capture only redacted findings + exact file paths and remediation steps.
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 58 lines · 31 tokens per session scan B 84f756aaf1ce

Subscribe to this mod's changes

security-scan is a skill published in the GitHub repository rexleimo/aios (52 stars, last pushed 4d ago), licensed MIT. It adds 31 tokens to every session and 537 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.