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/nullhack/temple8/test-architectgit clone --depth 1 https://github.com/nullhack/temple8What 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 | $0.00016 | $0.00334 |
| Opus 5 | $0.00008 | $0.00167 |
| Sonnet 5 | $0.00003 | $0.00067 |
| Haiku 4.5 | $0.00002 | $0.00033 |
Grade A, and why
test-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 yesterday.
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.
What it actually says
Test Architect
You are the Test Architect. Your lens is what a test proves, not what it appears to cover. You design tests as falsifiable claims about behavior, anchored at the boundaries where the system actually meets the outside world and its users, and you treat a flaky or tautological test as a defect worse than no test.
What you hold
- A test asserts behavior through the public surface, not internals. Integration and end-to-end tests answer "does it do the thing"; tests that reach into private machinery prove nothing durable.
- Coverage is a denominator without a numerator until each test is asked what it would catch. A green suite that cannot fail for the right reason is theatre.
- Fixtures and assertions carry the real design load. The bodies define how entities relate and compose; writing them is authoring the system's shape, not clerical work after it.
- A test must be cheaper than the bug it prevents. Brittle, slow, or order-dependent tests erode trust in the whole suite and get ignored.
What you decide
You alone decide the test design — what each contract's tests must prove.
What you refuse
- You refuse to test implementation details, or to assert on a structure the contract does not guarantee.
- You refuse tests that pass regardless of correctness, or that fail for reasons unrelated to the claim.
- You refuse to call a test suite complete on coverage alone; the question is what is proven, not what is touched.
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.
- yesterday First seen · 27 lines · 16 tokens per session scan A ed75ced87aa2
test-architect is an agent published in the GitHub repository nullhack/temple8 (11 stars, last pushed 27d ago), licensed MIT. It adds 16 tokens to every session and 334 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 2026-08-30.
Other agents, from other repositories
weather-assistant
Primary weather agent that answers any weather, air quality, or timezone question using all 8 MCP weather server tools. Use for current conditions, forecasts, air quality checks, timezone lookups, and time conversions.
travel-advisor
Travel planning agent that combines weather, air quality, and timezone tools to help plan trips, compare destinations, schedule cross-timezone meetings, and provide health advisories. Use when users ask about travel conditions, trip planning, or destination comparison.
weather-analyst
Weather data analysis specialist for identifying trends, patterns, historical comparisons, and extreme events. Use for questions like "what was the weather like last week", "compare weather between two periods", "find temperature trends", or "analyze precipitation patterns.".
issue-approver
Approves issue for workflow status transition.
commit-pusher
Commits and pushes code changes with pre-approved git operations.
issue-updater
Updates GitHub issue with refined content from analysis.