diff options
| author | main <main@swarm.moe> | 2026-03-20 00:33:08 -0400 |
|---|---|---|
| committer | main <main@swarm.moe> | 2026-03-20 00:33:08 -0400 |
| commit | ce41a229dcd57f9a2c35359fe77d9f54f603e985 (patch) | |
| tree | 1d42649c5711bf83cb738c40d83b81cbe7b08238 /assets/codex-skills/frontier-loop | |
| parent | 5047a141c45d18ef23ddd369fb262ecac867da11 (diff) | |
| download | fidget_spinner-ce41a229dcd57f9a2c35359fe77d9f54f603e985.zip | |
Refound ontology around hypotheses and experiments
Diffstat (limited to 'assets/codex-skills/frontier-loop')
| -rw-r--r-- | assets/codex-skills/frontier-loop/SKILL.md | 7 |
1 files changed, 4 insertions, 3 deletions
diff --git a/assets/codex-skills/frontier-loop/SKILL.md b/assets/codex-skills/frontier-loop/SKILL.md index b3ea44d..e8f1b94 100644 --- a/assets/codex-skills/frontier-loop/SKILL.md +++ b/assets/codex-skills/frontier-loop/SKILL.md @@ -62,7 +62,7 @@ ASSUME YOU ARE RUNNING OVERNIGHT. 2. Study existing evidence from `fidget-spinner`. 3. Search outward if the local frontier looks exhausted or you are starting to take unambitious strides. 4. Form a strong, falsifiable hypothesis. -5. Make the change. +5. Record the hypothesis and open the experiment. 6. Measure it. 7. If the result is surprising, noisy, or broken, debug the implementation and rerun only enough to understand the outcome. @@ -99,11 +99,12 @@ Every real experiment must leave an auditable record in `fidget-spinner`. If something matters to the frontier, put it in the DAG. -Use off-path records liberally for enabling work, side investigations, and dead +Use off-path records liberally for source capture, side investigations, and dead ends. When a line becomes a real measured experiment, close it through the proper -`fidget-spinner` path instead of improvising a chain of half-recorded steps. +`fidget-spinner` path instead of improvising a chain of half-recorded steps: +`hypothesis.record` -> `experiment.open` -> `experiment.close`. ## Resume Discipline |