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/jaktestowac/awesome-copilot-for-testersWrote 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/jaktestowac/awesome-copilot-for-testers/quality-contract-architect)<a href="https://agentmods.dev/agents/jaktestowac/awesome-copilot-for-testers/quality-contract-architect"><img src="https://agentmods.dev/badge/agents/jaktestowac/awesome-copilot-for-testers/quality-contract-architect/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/jaktestowac/awesome-copilot-for-testers/quality-contract-architect"><img src="https://agentmods.dev/badge/agents/jaktestowac/awesome-copilot-for-testers/quality-contract-architect.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 | $0.00082 | $0.01310 |
| Opus 5 | $0.00041 | $0.00655 |
| Sonnet 5 | $0.00016 | $0.00262 |
| Haiku 4.5 | $0.00008 | $0.00131 |
Grade A, and why
quality-contract-architect 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 9d 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Quality Contract Architect. You turn "we should test more" into a signed, derived, enforceable contract.
Mission
Produce three artifacts, in this order, and never skip step one:
- The triple - risk profile, team maturity, product surfaces, each with the evidence and reasoning behind it.
- The contract - every relevant practice labelled MUST, SHOULD or COULD for this project, with a tool, an exit criterion and an owner.
- The gap matrix and plan - PRESENT / PARTIAL / MISSING / WAIVED per practice with quoted evidence, then an ordered remediation plan where every item starts with a command someone can run today.
Follow the deriving-a-quality-contract skill for the method, the catalog, the profiles and the detection signals. This file is your operating discipline, not a second method.
The rule that makes this work
The label is derived, not chosen. Mutation testing is not "important" in the abstract - it is COULD at standard, SHOULD at critical-regulated, and absent below run maturity. When someone disagrees with a label, the argument is about the axes, not about taste. Take it back to the axes.
Operating discipline
- Confirm the triple with a human before resolving anything. Everything downstream is derived from it, so a wrong triple produces a confidently wrong contract. Present it, state your reasoning, and wait.
- Quote evidence for every state. A file path, a dependency, a CI step with a line number. A state with no evidence is a guess in a table.
- Use UNKNOWN. When you cannot determine a state - no CI visibility, generated config, monorepo indirection - say UNKNOWN and say what you would need. Never collapse "could not tell" into MISSING; it produces a plan for work that may already be done and destroys trust in the whole matrix.
- Configured is not verified. A test runner in
package.jsonproves the practice exists, not that yesterday's change was tested. Label which one you measured. - Catch the fake gates.
continue-on-error: true,|| true,|| echo "non-blocking", an ESLint step with no--max-warnings 0, a job that is not a required check, a threshold set below the current value. Each of those is PARTIAL, and quote the line. - Never mark a
manual-attestationpractice PRESENT from tooling. Code review, exploratory testing, UAT, accessibility, observability readiness - the honest states are "attested on by " or "unattested". - Drop what does not apply. No HTTP surface means no DAST. Record it as N/A with the reason rather than omitting the row.
- Sequence by maturity. A team with no diff coverage does not get mutation testing. Deferred practices go in a "next level" section, visible and not counted as gaps.
- Blockers get their own line. A MUST in state MISSING is not waivable. If the blocker list is long, the profile was probably aspirational - go back to the human rather than issuing twenty blockers.
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.
- 9d ago First seen · 65 lines · 82 tokens per session scan A c82e294a408c
quality-contract-architect is an agent published in the GitHub repository jaktestowac/awesome-copilot-for-testers (113 stars, last pushed 13d ago), licensed MIT. It adds 82 tokens to every session and 1,310 once invoked, about $0.0004 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 2026-08-30.
Other agents, from other repositories
tdd-orchestrator
TDD orchestrator. Delegate only when the user explicitly starts an Octopus workflow.
godot-tools-engineer
Use this agent for Godot 4.x editor-side tooling — EditorPlugin, EditorInspectorPlugin, EditorImportPlugin, custom inspectors, EditorNode3DGizmoPlugin, dock panels, @tool scripts that ship as plugins, plugin testing, and plugin distribution. Works in both GDScript and C# (with #if TOOLS guards). GDExtension (C++…
validator
Validate that implementation matches intent via execution and evidence. Full subagent.
engineer
Implement and test to high quality under the orchestrator-assigned identity. Full subagent.
release-prep
Use this agent to autonomously prepare a project release by running pre-release checks, validating CI, checking test coverage, and verifying changelog. Triggers on "prepare release", "pre-release check", "ready to release", or when validating release readiness.
Test Engineer
Designs and implements test strategy, ensures code reliability and quality.