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.
npx skills add UnboundCompute/security-agent-skills --skill hunting-os-command-injectiongit clone --depth 1 https://github.com/UnboundCompute/security-agent-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/unboundcompute/security-agent-skills/hunting-os-command-injection)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/hunting-os-command-injection"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-os-command-injection/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/hunting-os-command-injection"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-os-command-injection.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00177 | $0.02017 |
| Opus 5.5 | $0.00071 | $0.00807 |
| Sonnet 5.5 | $0.00035 | $0.00403 |
| Haiku 4.5 | $0.00018 | $0.00202 |
Grade A, and why
hunting-os-command-injection scanned grade A with 0 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 1mo 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hunting OS command injection: when input crosses into a shell
Command injection happens when untrusted input reaches a process-spawning API in a way that lets a shell interpret it. A shell treats characters like the semicolon, the pipe, the ampersand, backticks, and dollar parentheses as syntax, so an input that carries them can chain, substitute, or replace the intended command. The single fact that decides whether the class exists is whether a shell is in the loop: a call that composes a string and hands it to a system shell is injectable, while a direct exec that passes a fixed program and a separate argument vector, with no shell, is not, because the arguments are delivered as data. You find command injection by locating every spawn, deciding whether it goes through a shell, and tracing whether untrusted input reaches the command portion of a shell-interpreted call.
When to use
- Code runs an external command built from, or including, request or stored input.
- A spawn API is called in shell mode, or with a single composed string rather than an argument vector.
- Input influences the environment, the working directory, or the PATH of a spawned process.
Scope check
Test command injection only against systems you own or are authorized to assess, on non-production hosts, because a confirmed case runs commands on the server. Prove the class with an inert, side-effect-free command (an echo or a fixed delay) rather than anything destructive, and stay within the authorized host. If you can't name the authorization, stop.
The loop
- Establish whether a shell is in the loop first. For each spawn, determine whether it invokes a shell (a system-style call, a shell-mode flag, or a single composed command string) or is a direct exec with a fixed program and a separate argument vector. This is the false-positive killer: a shell-less exec does not interpret metacharacters, so untrusted arguments there are not command injection, even though they may be argument injection, a separate concern. Name the spawn mode before judging the input.
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.
- 1mo ago First seen · 132 lines · 177 tokens per session scan A d783fe805f01
hunting-os-command-injection is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 177 tokens to every session and 2,017 once invoked, about $0.0007 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-05.
Other skills, from other repositories
prodcheck-review
Review this codebase against the prodcheck pre-production checklists — security, performance, scale, integrations, post-launch readiness and people. Use when asked to check whether a project is ready to ship, to audit an area before launch, or to work through a specific checklist. Produces evidence with file:line…
integrate-arcjet-guard-genkit
Integrate Arcjet security into a Genkit JS agent using @arcjet/guard — wrap ai.defineTool, put guardMiddleware on generate({ use }) for unwrapped / MCP / filesystem tools, and read a caller-owned id from generate({ context }). Use when asked to add Arcjet to genkit, rate limit its tools, screen inbound messages, or…
integrate-arcjet-guard-tanstack-ai
Integrate Arcjet security into a TanStack AI chat() app using @arcjet/guard — put guardMiddleware first on chat({ middleware }) so onBeforeToolCall gates tools, and read a caller-owned id from chat({ context }). Use when asked to add Arcjet to TanStack AI, rate limit its tools, screen inbound messages, or block prompt…
integrate-arcjet-guard-claude-managed-agents
Integrate Arcjet security into Claude Managed Agents (hosted REST+SSE, beta managed-agents-2026-04-01) using @arcjet/guard — screen user.message / initialevents before sessions.events.send, and gate custom tools on agent.customtooluse. Use when asked to add Arcjet to Claude Managed Agents, rate limit custom tools, or…
linear-claude-skill
Manage Linear issues, projects, and teams.
varlock-claude-skill
Secure environment variable management ensuring secrets are never exposed in Claude sessions, terminals, logs, or git commits.