Skip to content

MiniMax-AI/mcode-stack / stacks/ronny/skills/harness-experiment

harness-experiment

Unverified

Run a harness change as a controlled experiment, with a small end-to-end eval, one change at a time, repeated runs, and looking at flipped tasks, request count, tokens and cache hits, not just the average score. Use before and after changing prompts, tools, skills, subagent layout, compaction or other agent scaffolding, or when asked whether a change helped.

Harness Experiment

Harness engineering moves forward by experiment. The principle: measure first, change one thing at a time, keep what survives testing.

Companion to MCode Engineering Notes 01, 02 and 05.

1. Before changing anything

  1. State the hypothesis in one sentence: what you change, what you expect to improve, and what might get worse.
  2. Pick or build a small end-to-end eval. A handful of golden tasks is enough to start. Include tasks for every chain the agent runs in production and every rule it must not break. The checks must not be written by the agent under test; tests an agent writes for itself share its blind spots.
  3. Pin everything else. Same model version, same settings, same tool set, same repository state. If the model itself changes, treat that upgrade as the one change and rerun the same eval.
  4. Record a baseline. Run each task several times; agent runs are noisy and one run can hide a failure.

2. Make exactly one change

One prompt edit, one tool moved, one subagent merged, one threshold changed. If you need several, run several experiments.

3. Measure the same things on both sides

For baseline and candidate, record:

MetricWhy
Pass rate per task (over repeated runs)Quality, with noise visible
Flipped tasks (pass to fail, fail to pass)Averages hide regressions
Model requests per taskMany wins come from fewer round trips
Input and output tokensCost and context pressure
Cache hit rateA change can save tokens but break caching
Wall-clock timeLatency the user feels

Only report numbers you actually measured, with run counts.

4. Read the result

  • Open every flipped task and read the trajectory, not just the final answer (see trajectory-review). Find out why it flipped.
  • A change that keeps the score flat but cuts requests or tokens can be a win. A change that raises the average while breaking a must-not-break rule is not.
  • If the evidence is mixed, keep the simpler version.
  • Write the result down: hypothesis, change, numbers, flipped tasks, decision. Next to the code (a PR description or an experiments log) is best.

5. Re-check after model upgrades

Scaffolding built around a model's weak spots ages fastest. After each model upgrade, rerun the eval and ask which parts of the harness still earn their keep.

Output

A short experiment record: hypothesis, the single change, eval set and run count, the metric table for baseline and candidate, flipped tasks with causes, and the decision (keep / revert / iterate).

reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 463kUpdated Claude CodeCodexVerifiedInstall
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 463kUpdated Claude CodeCodexVerifiedInstall
discernment-nudgeanthropics/skillsAfter you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.Stars: 195kUpdated Claude CodeCodexVerifiedInstall
internal-commsanthropics/skillsA set of resources to help me write all kinds of internal communications, using the formats that my company likes to use. Claude should use this skill whenever asked to write some sort of internal communications (status reports, leadership updates, 3P updates, company newsletters, FAQs, incident reports, project updates, etc.).Stars: 195kUpdated Claude CodeCodexVerifiedInstall