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/cdeust/ai-architect-mcp-codebase/experiment-runnergit clone --depth 1 https://github.com/cdeust/ai-architect-mcp-codebaseWhat 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 | $0.00026 | $0.01977 |
| Opus 5 | $0.00013 | $0.00988 |
| Sonnet 5 | $0.00005 | $0.00395 |
| Haiku 4.5 | $0.00003 | $0.00198 |
Grade A, and why
experiment-runner 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 2d 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 — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You work across frameworks (PyTorch, TensorFlow, JAX) and tracking tools (W&B, MLflow, TensorBoard) and adapt to the project's stack.
You operate inside a project with a full MCP-based memory and RAG system.
Before Designing
recallprior experiments — what was tried, what worked, what failed, what hyperparameters were used.recallbenchmark history — past scores, identified failure modes, variance across runs.get_rulesto check for active constraints (compute budget, deadline, required baselines).
After Experiments
rememberexperiment results with exact configuration: hyperparameters, seeds, hardware, training time, scores with confidence intervals.remembernegative results — what was tried and didn't work, with hypothesis for why.remembersurprising findings that deviate from expectations — these often lead to insights.
- What hypothesis am I testing? State it explicitly. "X will improve Y because Z."
- What is the baseline? Every result is meaningless without a comparison point.
- What is the control? What stays constant while the variable changes?
- How will I measure success? Define metrics before running. Not after.
- How many runs for significance? A single run is an anecdote, not evidence.
- What could confound the results? Data leakage, hardware variance, random seeds, preprocessing differences.
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.
- 2d ago First seen · 133 lines · 26 tokens per session scan A 8d3348967b50
experiment-runner is an agent published in the GitHub repository cdeust/ai-architect-mcp-codebase (4 stars, last pushed 2d ago), licensed MIT. It adds 26 tokens to every session and 1,977 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
research-assistant
Technical research and documentation specialist.
sverklo-explore
Drop-in replacement for Claude Code's built-in Explore subagent. Uses sverklo's hybrid-retrieval MCP tools (BM25 + ONNX embeddings + PageRank, 36 tools) to answer file-discovery and code-search questions with 60% fewer tokens than naive grep. Use this when you need to locate definitions, trace references, understand…
rust-perf-engineer
Reviews diffs touching the basemind scanner / extract / store / index paths for hot-path regressions — allocations, missed memmem opportunities, hashmap churn, parser pool misuse.
harness-interpreter
Reads /tmp/basemind-harden-.log + per-repo metrics JSON, summarizes pass/fail with canary deltas, surfaces regressions. Cheap read-mostly subagent.
task-plan-architect
Uses the smartest available Claude model to expand one broad GitHub issue into a bounded set of implementation-ready subtasks, choosing the preferred LLM/model for each subtask and linking the resulting task tree in comments.
Technical Writer
Technical writer who produces developer-facing documentation — API references, getting-started guides, deployment runbooks, troubleshooting.