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 reviewing-detection-rules-for-evasiongit 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/reviewing-detection-rules-for-evasion)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/reviewing-detection-rules-for-evasion"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/reviewing-detection-rules-for-evasion/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/reviewing-detection-rules-for-evasion"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/reviewing-detection-rules-for-evasion.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 | $0.00176 | $0.01923 |
| Opus 5.5 | $0.00070 | $0.00769 |
| Sonnet 5.5 | $0.00035 | $0.00385 |
| Haiku 4.5 | $0.00018 | $0.00192 |
Grade A, and why
reviewing-detection-rules-for-evasion 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.
Reviewing detection rules for evasion: whether the rule survives an attacker who reads it
A detection rule is a guard, and like any guard it is only as strong as the exactness of the thing it checks against what the attacker can vary. Detection content is often written to match one observed spelling of an attack: a specific command line, a known tool name, a particular flag. An attacker who can read the rule, or simply reason about it, expresses the same action a different way and walks past. This review treats a rule the way an attacker treats a filter: find the field it trusts, find the variant it does not cover, and produce an event that does the malicious thing without tripping the match. The attacker-controlled field is the source, the rule's match logic is the guard, and a bypass is the bug.
When to use
- You maintain detection-as-code rules and want to know which are brittle before an attacker does.
- Rules match on attacker-controlled fields: command lines, arguments, user-agents, paths, names.
- You are reviewing or threat-modeling existing detections, not writing new ones.
Scope check
Test evasion only against detections, telemetry, and systems you own or are authorized to assess, and generate malicious-looking events only where that is expected and coordinated. A realistic bypass event can trigger a real response. If you can't name the authorization, stop.
The loop
-
Inventory each rule and the field it trusts. For every detection rule, identify the fields it matches on and which of those an attacker controls freely: a process command line and its arguments, environment values, a user-agent, a file path or name. A rule keyed on an attacker-controlled field is only as strong as how exactly it pins that field against every equivalent form.
-
Test literal and encoding variants. For each match, ask whether the same malicious action can be expressed a way the rule does not cover: alternate casing, quoting, extra whitespace, an equivalent path or name form, reordered flags, an encoded argument, or a different tool that has the same effect. A rule that matches one literal spelling misses every synonym of it.
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 · 176 tokens per session scan A 591875908a6a
reviewing-detection-rules-for-evasion is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 176 tokens to every session and 1,923 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.