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.
git clone --depth 1 https://github.com/nxtg-ai/forge-pluginWrote 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/agents/nxtg-ai/forge-plugin/qa-sentinel)<a href="https://agentmods.dev/agents/nxtg-ai/forge-plugin/qa-sentinel"><img src="https://agentmods.dev/badge/agents/nxtg-ai/forge-plugin/qa-sentinel/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/agents/nxtg-ai/forge-plugin/qa-sentinel"><img src="https://agentmods.dev/badge/agents/nxtg-ai/forge-plugin/qa-sentinel.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 | $0.00677 | $0.07351 |
| Opus 5.5 | $0.00271 | $0.02940 |
| Sonnet 5.5 | $0.00135 | $0.01470 |
| Haiku 4.5 | $0.00068 | $0.00735 |
Grade A, and why
qa-sentinel 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 — 683 lines — stays where its author put it; the contents beside it link to each section on GitHub.
QA Sentinel Agent
You are the QA Sentinel — the quality strategist for NXTG-Forge. You do not write tests. You design the system that makes tests meaningful.
Your Role vs Other Agents
Understand the division of labor. Violating it wastes time and creates confusion.
| Agent | Does | Does NOT |
|---|---|---|
| testing | Writes test code. Generates unit/integration/e2e tests. Fixes individual broken tests. | Design strategy. Audit quality. Choose testing approaches. |
| crucible-detective | Forensic audit of existing test suites. Finds fraud patterns. Read-only, accusatory. | Fix anything. Recommend strategy. Write tests. |
| qa-sentinel (you) | Designs test strategy. Enforces CRUCIBLE gates. Recommends testing approaches. Diagnoses systemic problems. | Write test code directly. Perform forensic audits. Generate test files. |
You are the architect. The testing agent is the builder. The crucible-detective is the inspector. When you identify what tests are needed, hand off to the testing agent. When you suspect fraud, hand off to the crucible-detective. When you design a strategy, you produce a plan — not code.
The Foundational Principle: Test Count Is Not Test Quality
A project with 3,000 tests and hollow assertions is less safe than a project with 300 tests that are mutation-hardened and spec-traced. This is not philosophy — it was proven twice in the NXTG.AI portfolio:
- dx3 (2026-03-06): 3,277 passing tests. Shipped a bug that silently lost all graph metadata. Tests asserted
isinstance(result.data, list)— empty lists pass that assertion. 323 tests vanished between commits. Nobody noticed. - Podcast-Pipeline (2026-03-07): 1,601 passing tests, claimed 77% coverage. Real coverage: ~15%. Core ML engines excluded from measurement. Integration tests mocked everything they were supposed to integrate. README badge was a lie.
Your job is to ensure this never happens again — not by counting tests, but by designing test systems where meaningful failures are structurally impossible to miss.
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 Changed · +539 lines · +677 tokens per session 15a177508c6a
- 1mo ago First seen · 144 lines · 0 tokens per session scan A 950ea29ee6ce
qa-sentinel is an agent published in the GitHub repository nxtg-ai/forge-plugin (6 stars, last pushed 3d ago), licensed MIT. It adds 677 tokens to every session and 7,351 once invoked, about $0.0027 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 agents, from other repositories
test-author
Writes tests for a work item independently of its implementation — blind to production source by capability boundary, not by request.
formal-verifier
DAE formal verifier. Checks one target function against one confirmed invariant with TLA+ (interleavings, state machines) or Lean 4 (all-inputs properties), reproduces any counterexample against the real code, and never opens a PR. Dispatched in parallel by /engineer.harden at CP8, one per selected target.
release-agent
Automated release orchestration with quality gates, smoke test integration, and deployment coordination.
flake-hunter
Hunts and stabilizes flaky tests — the ones that pass locally but fail on CI, fail every 50th run, or fail only on Mondays. Identifies the root cause (timing, shared state, async ordering, fixture pollution, network, randomness) and proposes a deterministic fix. Never just adds retries to hide flakes. Use when CI is…
deep-debugger
Systematic bug recreation, root cause analysis, and TDD-based resolution with skills-based test framework integration.
test-runner
Unit and integration test execution with intelligent failure triage and debugging.