Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/danieleteti/delphi-ai-skillsnpx agentmods add skills/danieleteti/delphi-ai-skills/dmvcframework-securityWrote 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/danieleteti/delphi-ai-skills/dmvcframework-security)<a href="https://agentmods.dev/skills/danieleteti/delphi-ai-skills/dmvcframework-security"><img src="https://agentmods.dev/badge/skills/danieleteti/delphi-ai-skills/dmvcframework-security/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/danieleteti/delphi-ai-skills/dmvcframework-security"><img src="https://agentmods.dev/badge/skills/danieleteti/delphi-ai-skills/dmvcframework-security.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 | $0.00112 | $0.05211 |
| Opus 5.5 | $0.00045 | $0.02084 |
| Sonnet 5.5 | $0.00022 | $0.01042 |
| Haiku 4.5 | $0.00011 | $0.00521 |
Grade B, and why
dmvcframework-security scanned grade B with 1 finding 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 23d 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.
Cloud metadata endpointmediumServer-side request forgery
One request to 169.254.169.254 can return temporary IAM credentials.
`169.254.169.254` (cloud metadata), your internal network. If a feature needs to fetch a URL, allow-list the Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
Copies of this mod
1 near-identical copy found in the catalogue:
- dmvcframework-security — 100% identical, 105 lines differ
How it starts
The opening of the file, as written. The whole thing — 432 lines — stays where its author put it; the contents beside it link to each section on GitHub.
DMVCFramework — Secure coding reference
This is a reference for the other DMVCFramework skills. Whenever you write an endpoint that accepts input from a client — a controller action, a minimal-API handler, a web page, a filter — the controls below apply. They are not optional extras.
Adapted from the OWASP-style VibeSec secure-coding guide
(Apache-2.0), rewritten against the real DMVCFramework API: every class, attribute and filter named here was
verified in sources/.
The rule that catches most of it: never trust anything that arrived over the wire — not the body, not the query string, not a header, not a cookie, not a filename, not a URL, not a JWT you have not verified.
When in doubt about an API — verify it, never guess
Never invent an identifier or answer from memory. If you need a signature this skill does not cover, ask the
user for the path to their DelphiMVCFramework checkout and read sources/ (and the matching samples/
project); failing that, read the official repository —
sources ·
samples.
If you still cannot verify it, say so. Where a skill and a sample disagree, the sample wins.
1. Access control — the #1 cause of real breaches
Authentication says who you are. Authorization says what you may touch. Do not confuse them: a valid JWT is not permission to read row 42.
The IDOR trap. This is the single most common serious bug in a CRUD API:
// BROKEN — any authenticated user can read anyone's order
function TOrdersController.GetOrder(const ID: Integer): IMVCResponse;
begin
Result := OKResponse(TMVCActiveRecord.GetByPK<TOrder>(ID));
end;
The ID comes from the URL. Nothing ties it to the caller. Scope every lookup to the authenticated
principal, in the query — not with an if after the fact:
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.
- 23d ago Changed · +9 lines 861cd7569c3e
- 24d ago Changed · +56 lines ef728b1f3b57
- 1mo ago First seen · 367 lines · 112 tokens per session scan B 0888a42c4ca3
dmvcframework-security is a skill published in the GitHub repository danieleteti/delphi-ai-skills (20 stars, last pushed 19d ago), licensed Apache-2.0. It adds 112 tokens to every session and 5,211 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it B with 1 finding (cloud metadata endpoint). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
intuitive-doc
Maintain README.md, ARCHITECTURE.md, STATUS.md, and docs/human/ when they drift or need cleanup.
intuitive-shape
Shape unsettled product ideas into bounded bets before planning; excludes settled fixes and approved execution.
intuitive-reduce-entropy
Find and rank material simplification candidates when the cleanup target or plan risks are unclear.
research
Investigate a bounded question using multiple sources or a cited report; answer single facts and codebase questions directly.
agent-planning-loop
Review unsettled plans with independent scouts when the user requests multi-agent planning or alignment.
intuitive-flow
Execute an approved plan or preflight contract, or a tiny bounded change; route discovery and refactors to their owners.