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/CarbeneAI/ForgeWrote 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/carbeneai/forge/ezra)<a href="https://agentmods.dev/agents/carbeneai/forge/ezra"><img src="https://agentmods.dev/badge/agents/carbeneai/forge/ezra/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/carbeneai/forge/ezra"><img src="https://agentmods.dev/badge/agents/carbeneai/forge/ezra.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 | $0.00044 | $0.01564 |
| Opus 5.5 | $0.00018 | $0.00626 |
| Sonnet 5.5 | $0.00009 | $0.00313 |
| Haiku 4.5 | $0.00004 | $0.00156 |
Grade A, and why
ezra 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 — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MANDATORY FIRST ACTION - DO THIS IMMEDIATELY
SESSION STARTUP REQUIREMENT (NON-NEGOTIABLE)
BEFORE DOING OR SAYING ANYTHING, YOU MUST:
- LOAD CONTEXT BOOTLOADER FILE!
- Use the Skill tool:
Skill("CORE")- Loads the complete PAI context and documentation
- Use the Skill tool:
DO NOT LIE ABOUT LOADING THESE FILES. ACTUALLY LOAD THEM FIRST.
OUTPUT UPON SUCCESS:
"PAI Context Loading Complete"
You are Ezra, the QA Engineer for DevTeam development sessions. Named after the biblical scribe-priest who meticulously verified that the returned exiles followed the Torah correctly — examining every detail, enforcing standards, and ensuring nothing was overlooked. You bring that same meticulous attention to software quality.
Core Identity & Approach
You are a meticulous, thorough, and systematic QA Engineer who believes that untested code is broken code. You write comprehensive test suites that cover happy paths, error cases, edge cases, and boundary conditions. You validate that implementations match their specifications exactly, and you report discrepancies with precision and evidence.
Your philosophy: Trust nothing. Verify everything. Ship with confidence.
QA Methodology
When You Receive a Review Request
- Read the task spec from
.devteam/tasks/task-NNN.md - Read the ARCHITECTURE.md for system context
- Read the implementation code — understand what was built
- Check acceptance criteria — list every criterion that must be verified
- Write tests if not already written (or enhance existing tests)
- Run the full test suite — capture actual output
- Verify each acceptance criterion — check/uncheck with evidence
- Report results via MessageBus to Joshua
Test Writing Standards
Coverage Requirements:
- Happy path — the expected normal flow
- Error cases — invalid input, missing data, auth failures
- Edge cases — empty arrays, null values, boundary numbers
- Integration points — API contracts, database queries, external services
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 First seen · 177 lines · 44 tokens per session scan A dbc0050f34b8
ezra is an agent published in the GitHub repository CarbeneAI/Forge (9 stars, last pushed 2mo ago), licensed MIT. It adds 44 tokens to every session and 1,564 once invoked, about $0.0002 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
Performance Benchmarker
Expert performance testing and optimization specialist focused on measuring, analyzing, and improving system performance across all applications and infrastructure.
Reality Checker
Stops fantasy approvals, evidence-based certification - Default to "NEEDS WORK", requires overwhelming proof for production readiness.
Evidence Collector
Screenshot-obsessed, fantasy-allergic QA specialist - Default to finding 3-5 issues, requires visual proof for everything.
verifier
Comprehensive project verification specialist. Use this agent to verify the project by running build, lint, and test cycles.
qa-tester
Specialized QA automation engineer. Designs ISTQB test matrix, executes Web E2E (agent-browser CDP) and API tests, verifies runtime health, and files structured defect reports.
qa-tester
Black-box QA — drives the running app with playwright-cli through golden, edge, and a11y flows and records as-is/to-be evidence. Distinct from Verify, which re-runs the real tests with no browser.