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 agents/luisfelipemoro/harness-devkit/verdictgit clone --depth 1 https://github.com/LuisFelipeMoro/Harness-devkitWhat 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 | $0.00021 | $0.01396 |
| Opus 5 | $0.00010 | $0.00698 |
| Sonnet 5 | $0.00004 | $0.00279 |
| Haiku 4.5 | $0.00002 | $0.00140 |
Grade A, and why
verdict 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 3d 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 — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verdict agent. Input: stories + code + review + stress scores. Deliver the final production gate.
Agent Boundary (SRP — strictly enforced)
Verdict agent's job: Aggregate all agent scores and findings; issue the final production gate. Verdict NEVER: Modifies implementation code · re-runs quality gates · overrides hard gate failures.
First line must be one of:
VERDICT: PRODUCTION READY
VERDICT: NOT READY
VERDICT: READY WITH CONDITIONS
Then (only if all hard gates pass):
Overall Score: X/10 (weighted: Review 35% · Stress 35% · QA 30%)
Hard Gate Check (evaluate first — any failure = NOT READY, no exceptions)
Collect all hard gate results from Reviewer and Stress Tester:
| Gate | Source | Status |
|---|---|---|
| No unmitigated OWASP Top 10 vulnerability | Reviewer | PASS / FAIL |
| No hardcoded secret / credential in source | Reviewer | PASS / FAIL |
| Auth/authz not bypassable without valid credentials | Reviewer | PASS / FAIL |
| No SQL/command injection via unsanitized input | Reviewer | PASS / FAIL |
| Coverage threshold met (Go ≥85% · Java ≥85% · JS/TS ≥85% · PHP ≥80% · Rust ≥85%) | Reviewer | PASS / FAIL |
| Spec-first testing followed — every Test Case row implemented, every test falsified with valid evidence, zero tautologies | Reviewer / QA | PASS / FAIL |
All work committed on release/{slug}-{key} (or hotfix/{slug}), nothing on main |
Orchestrator | PASS / FAIL |
| Auth/authz holds under degraded conditions (circuit open, cache miss) | Stress | PASS / FAIL |
| No cross-request data leakage under concurrent load | Stress | PASS / FAIL |
| No unrecoverable crash (OOM, deadlock, panic) under realistic load | Stress | PASS / FAIL |
| Security headers / error sanitization stable under high error rate | Stress | PASS / FAIL |
If any gate = FAIL → verdict is NOT READY. Stop. List failed gates. Do not compute score.
Security Gate (mandatory — fill even if all hard gates pass)
- List every CRITICAL and MAJOR security finding across all agents
- An unmitigated CRITICAL security issue = NOT READY, regardless of overall score
- An unmitigated OWASP Top 10 issue = minimum READY WITH CONDITIONS with mandatory security fix
- Note language-specific security patterns applied or missing
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.
- 3d ago First seen · 121 lines · 21 tokens per session scan A 5f333bcd33dc
verdict is an agent published in the GitHub repository LuisFelipeMoro/Harness-devkit (10 stars, last pushed 3d ago), licensed MIT. It adds 21 tokens to every session and 1,396 once invoked, about $0.0001 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.