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 Quality-Max/free-qa-skills --skill agentic-app-risk-reviewgit clone --depth 1 https://github.com/Quality-Max/free-qa-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/quality-max/free-qa-skills/agentic-app-risk-review)<a href="https://agentmods.dev/skills/quality-max/free-qa-skills/agentic-app-risk-review"><img src="https://agentmods.dev/badge/skills/quality-max/free-qa-skills/agentic-app-risk-review/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/quality-max/free-qa-skills/agentic-app-risk-review"><img src="https://agentmods.dev/badge/skills/quality-max/free-qa-skills/agentic-app-risk-review.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 | $0.00060 | $0.00676 |
| Opus 5 | $0.00030 | $0.00338 |
| Sonnet 5 | $0.00012 | $0.00135 |
| Haiku 4.5 | $0.00006 | $0.00068 |
Grade A, and why
agentic-app-risk-review 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 11d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agentic App Risk Review
Find the security weaknesses in your own LLM or agent app before users — or attackers — do. Read-only: it flags risks, it doesn't change your code.
Prerequisites
- None. Pure Claude Code — reads your repository directly, no MCP required.
Trigger
- "Review my agent app for prompt injection"
- "Security review of my LLM integration"
- "Is my chatbot safe against jailbreaks / tool abuse?"
Workflow
- Locate the AI surface: where prompts are built, where the model is called, and where it can take actions (tool/function calls, shell, HTTP, DB, file writes, payments).
- Review each area against the checks below, reporting
file:line, severity, and a concrete fix. Treat all model input and tool output as untrusted.
What to flag
| Risk | What to look for |
|---|---|
| Prompt injection | Untrusted input (user text, web/RAG content, file contents) concatenated into the prompt without delineation or instruction-hardening |
| Indirect injection | Fetched pages, emails, or documents fed to the model that can carry instructions the agent then follows |
| Unsafe tool calls | Tools that run commands, write files, send money, or call APIs with no allowlist, schema validation, or argument sanitization |
| Excessive agency | Destructive/irreversible tools exposed with no human-in-the-loop confirmation |
| Unbounded loops | Agent/tool loops with no max-step or cost/timeout limit (DoS / runaway spend) |
| Output not validated | Model output used directly in shell, SQL, HTML, or eval without checking |
| PII / secret leakage | Secrets, keys, or personal data placed into prompts, tool args, or logs/traces |
| Over-broad system prompt | Secrets or privileged instructions in the system prompt that injection could exfiltrate |
| No rate / cost limits | Per-user or per-session limits absent on model and tool calls |
- Output a severity-ranked report:
## Agentic App Risk Review
**2 high, 1 medium**
### HIGH
- agent/tools.py:40 — shell tool runs model-provided strings with no allowlist
or validation (command injection via tool abuse). Restrict to a fixed command
set and validate arguments.
- chat/prompt.py:22 — user message concatenated straight into the system prompt
(prompt injection). Separate untrusted input from instructions and harden.
### MEDIUM
- agent/loop.py:15 — tool loop has no max-step limit (runaway spend / DoS).
Add a step and cost ceiling.
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.
- 11d ago First seen · 62 lines · 60 tokens per session scan A e41bb9e1268b
agentic-app-risk-review is a skill published in the GitHub repository Quality-Max/free-qa-skills (10 stars, last pushed 18d ago), licensed Apache-2.0. It adds 60 tokens to every session and 676 once invoked, about $0.0003 per session on Opus 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 2026-08-31.
Other skills, from other repositories
accessibility
Audit and improve web accessibility following WCAG 2.2 guidelines. Use when asked to "improve accessibility", "a11y audit", "WCAG compliance", "screen reader support", "keyboard navigation", or "make accessible".
performance
Optimize web performance for faster loading and better user experience. Use when asked to "speed up my site", "optimize performance", "reduce load time", "fix slow loading", "improve page speed", or "performance audit".
seo
Optimize for search engine visibility and ranking. Use when asked to "improve SEO", "optimize for search", "fix meta tags", "add structured data", "sitemap optimization", or "search engine optimization".
best-practices
Apply modern web development best practices for security, compatibility, and code quality. Use when asked to "apply best practices", "security audit", "modernize code", "code quality review", or "check for vulnerabilities".
web-quality-audit
Run an evidence-led web quality audit covering performance, accessibility, SEO, best practices, and agentic browsing. Use when asked to "audit my site", "review web quality", "run lighthouse audit", "check page quality", or "optimize my website".
core-web-vitals
Optimize Core Web Vitals (LCP, INP, CLS) for better page experience using field and lab evidence. Use when asked to "improve Core Web Vitals", "fix LCP", "reduce CLS", "optimize INP", "page experience optimization", or "fix layout shifts".