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 hunting-orm-and-query-builder-injectiongit 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/hunting-orm-and-query-builder-injection)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/hunting-orm-and-query-builder-injection"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-orm-and-query-builder-injection/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/hunting-orm-and-query-builder-injection"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-orm-and-query-builder-injection.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 | $0.00177 | $0.01979 |
| Opus 5 | $0.00088 | $0.00989 |
| Sonnet 5 | $0.00035 | $0.00396 |
| Haiku 4.5 | $0.00018 | $0.00198 |
Grade A, and why
hunting-orm-and-query-builder-injection 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 6d 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 — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hunting ORM and query-builder injection: when the abstraction is not the safeguard
Teams reach for an object-relational mapper or a query builder partly to avoid injection, then assume the abstraction handles it. It handles values: a bound parameter is safe no matter what it contains. It does not handle structure. The moment untrusted input becomes a column name, a sort direction, a raw fragment, or an operator inside a filter object, the placeholder machinery is bypassed and the query means something the developer did not write. Classic string-concatenation injection is largely gone from ORM codebases; the live risk moved to the escape hatches and the structured inputs. You find it by separating the calls that bind values from the calls that accept structure and tracing untrusted input into the latter.
When to use
- Data access goes through an object-relational mapper or query builder, not hand-written statements.
- Untrusted input reaches a raw-query method, a column, table, or sort argument, or a filter object.
- Request bodies are passed as filter or update objects into the data layer.
Scope check
Test injection only against databases and applications you own or are authorized to assess, on non-production data. Confirming an operator-injection read can expose records outside your test scope. If you can't name the authorization, stop.
The loop
-
Map the data-access sinks and separate them. Inventory where the mapper or builder is used, then split the calls into two groups: parameterized calls that bind values, and escape hatches that accept structure. The escape hatches are raw-query methods, methods that take a column, table, or order argument, and methods that accept a structured filter or update object. Only the second group can be injected.
-
Trace untrusted input into raw-query escape hatches. A raw method takes a query string or a fragment. If untrusted input is concatenated or interpolated into that string, it is classic injection even though the file is full of ORM calls. Parameter placeholders inside a raw fragment protect the values only; anything spliced in as text is live.
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.
- 6d ago First seen · 135 lines · 177 tokens per session scan A 357228a2e61d
hunting-orm-and-query-builder-injection is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 3d ago), licensed MIT. It adds 177 tokens to every session and 1,979 once invoked, about $0.0009 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 2026-09-05.
Other skills, from other repositories
prodcheck-review
Review this codebase against the prodcheck pre-production checklists — security, performance, scale, integrations and post-launch readiness. 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 citations…
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-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.
panguard
AI agent security platform — audit skills, scan for threats, and run 24/7 protection with 9,700+ detection rules.
security-guide
A security deployment guide for OpenClaw, with instructions in English and Chinese. It covers network exposure, container isolation, credentials, permissions, and audit logging.