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 JayRHa/AgentSkills --skill secure-code-reviewgit clone --depth 1 https://github.com/JayRHa/AgentSkillsWrote 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/jayrha/agentskills/secure-code-review)<a href="https://agentmods.dev/skills/jayrha/agentskills/secure-code-review"><img src="https://agentmods.dev/badge/skills/jayrha/agentskills/secure-code-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/jayrha/agentskills/secure-code-review"><img src="https://agentmods.dev/badge/skills/jayrha/agentskills/secure-code-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 | $0.00120 | $0.01070 |
| Opus 5.5 | $0.00048 | $0.00428 |
| Sonnet 5.5 | $0.00024 | $0.00214 |
| Haiku 4.5 | $0.00012 | $0.00107 |
Grade A, and why
secure-code-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 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 — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Secure Code Review
Overview
Review code through the eyes of an attacker. For each finding, name the vulnerability class, show how it's exploited, rate severity, and give a concrete fix. Scope to the diff/files provided; pull in just enough surrounding code (callers, trust boundaries) to judge exploitability.
Keywords: security review, secure code review, vulnerability, OWASP, injection, SQLi, XSS, command injection, authentication, authorization, IDOR, broken access control, crypto, secrets, hardcoded credentials, deserialization, SSRF, path traversal, input validation, CSRF.
Workflow
- Map trust boundaries. Identify where untrusted data enters: HTTP params/body/headers, file uploads, env, message queues, DB rows from other tenants, third-party APIs. Untrusted input is where most bugs live.
- Trace tainted data to sinks. Follow each untrusted input to dangerous sinks (SQL, shell, file paths, HTML output, deserializers, redirects, template engines). A flow from source → sink without sanitization is a finding.
- Review each vulnerability class using
references/vuln-patterns.md: injection, broken access control/authz, auth & session, crypto, secrets, deserialization, SSRF, path traversal, CSRF, security misconfig, sensitive-data exposure. - Check authorization on every sensitive operation. Authentication ≠ authorization. Confirm object-level checks (does this user own this record?) to catch IDOR / broken access control.
- Scan for low-hanging fruit with
scripts/secret_scan.py(hardcoded secrets, dangerous function calls). Triage hits — it's a starting point, not proof. - Rate severity (Critical/High/Medium/Low) by exploitability × impact, using
references/checklist.md. - Write findings with: class,
file:line, exploit scenario, severity, and a concrete fix (prefer a diff). End with a verdict and a prioritized fix list.
Decision Heuristics
- Assume all external input is hostile until validated/escaped at the sink.
- Escape at the sink, validate at the boundary — parameterized queries beat input filtering.
- Deny by default for access control; explicit allow per resource and action.
- Never trust client-side checks — they're UX, not security.
- Secrets in code = Critical, even in tests or comments, because of git history.
- If uncertain it's exploitable, mark "Possible:" and state what would confirm it.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 66 lines · 120 tokens per session scan A 7ef926cc896c
secure-code-review is a skill published in the GitHub repository JayRHa/AgentSkills (12 stars, last pushed 2mo ago), licensed MIT. It adds 120 tokens to every session and 1,070 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
code-reviewer
Use when reviewing pull requests, conducting code quality audits, or identifying security vulnerabilities. Invoke for PR reviews, code quality checks, refactoring suggestions.
claude-md-check
Use when the user asks to review, audit, improve, shorten or write a CLAUDE.md, CLAUDE.local.md, AGENTS.md or .claude/rules file, or says Claude Code keeps ignoring their instructions or forgetting project conventions. Checks the file against the repository it describes and against how memory files actually load, then…
developer-workflow
Use when the user is writing, reviewing, debugging, testing, or refactoring code, or wants help with commits and pull requests. Emphasizes reading existing code first, verifying changes, and never inventing APIs.
supreme-project-audit
Evidence-driven full-project audit skill for Product Engineers, AI Engineers, ML Engineers, LLM Engineers, LLM Architects, and AI Researchers. Enforces severity discipline (P0/P1/P2/P3 with objective criteria), explicit coverage maps (audited vs not-audited surface), threat modeling (STRIDE + OWASP LLM Top 10)…
merge-batch
Prepare and merge an owner-selected PR batch with contributor credit and exact-candidate admission. Use for a pasted PR list; combined focused validation is diagnostic, and every actual landing still requires its applicable checks and authorization.
make-pr-easy-to-review
Prepare PRs for review by cleaning noisy history, improving PR descriptions, and adding reviewer guidance without changing code behavior. Use for "make this easy to review", "tidy this PR", "clean up commits", or "annotate the diff".