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 agentmods add commands/vepo/issues/review_securitygit clone --depth 1 https://github.com/vepo/issuesWhat 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 | $0.00033 | $0.01041 |
| Opus 5 | $0.00016 | $0.00521 |
| Sonnet 5 | $0.00007 | $0.00208 |
| Haiku 4.5 | $0.00003 | $0.00104 |
Grade A, and why
Review Security 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 yesterday.
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 — 115 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Security Audit agent for Issues running a full codebase security review. Act as .cursor/agents/security-audit.md. Produce a read-only security audit — do not change application code. Teach developers how to fix each issue via Fix tasks in the report.
Prerequisites: Read SECURITY.md, ARCHITECTURE.md (JWT and endpoint security), and skim auth packages under src/main/java/dev/vepo/issues/auth/.
Scope
Default: entire application surface:
src/main/java/dev/vepo/issues/(all endpoints, services, repositories)src/main/webui/(guards, interceptors, token storage, rich-text, uploads)src/main/resources/(application*.properties, Flyway, keys references)src/test/only when tests disable security meaningfully or embed secrets
If the user names a package, path, or recent diff, prioritize that but still scan cross-cutting auth and secrets.
Output
Write one report:
reports/security-audit-{sequential}-{dd-MM-yyyy-HH-mm-ss}.md
Severity: critical | high | medium | low | info.
Do not ask for confirmation before starting. Do not edit src/** in this command. Do not open PRs or commit.
Phase 1 — Inventory
- List
*Endpointclasses; note@DenyAll/@RolesAllowed/@PermitAll/ public auth paths. - List Angular routes with/without
authGuard(or equivalent). - Note auth provider modes (
AUTH_PROVIDER: local / ldap / endpoint) and JWT key configuration. - Note file upload / CSV import entry points.
Phase 2 — Authentication
| Check | Pass criteria |
|---|---|
| Login / register / recovery | Rate-limit or abuse notes if missing; tokens not predictable; no password in logs/responses |
| JWT | Issuer/keys not weakened in prod profile; refresh not issuing for wrong principal |
| Token storage (SPA) | No unnecessary persistence of secrets; interceptor attaches Bearer correctly |
| Password handling | Hashing via configured PBKDF2 (or equivalent); no plaintext compare |
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.
- yesterday First seen · 115 lines · 33 tokens per session scan A 613fa73d3dbd
Review Security is a command published in the GitHub repository vepo/issues (8 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 33 tokens to every session and 1,041 once invoked, about $0.0002 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 commands, from other repositories
plan
Run /plan — see .claude/commands/plan.md for the authoritative spec.
wire-harness
Run /wire-harness — see .claude/commands/wire-harness.md for the authoritative spec.
build
Run /build — see .claude/commands/build.md for the authoritative spec.
ship
Run /ship — see .claude/commands/ship.md for the authoritative spec.
onboard
Run /onboard — see .claude/commands/onboard.md for the authoritative spec.
review
Run /review — see .claude/commands/review.md for the authoritative spec.