- refs/ -> references/, the folder name the Agent Skills spec uses and the one
Hermes skips when it walks for nested skills.
- rl and pinn declared name: rl and name: pinn, which are global names in a flat
skill namespace. Now ml-debug-rl and ml-debug-pinn. They also called themselves
sub-skills of 'ml-debugging', which is not this skill's name.
- Drop the dead link to SKILL_old.md. It moved into gitignored slop/, so the link
was broken for anyone who cloned.
- Route references/llm_judge_litreview.md, the one reference SKILL.md never named.
- Description leads with the trigger situations. Hermes truncates it to 57 chars.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
Headings were the practitioner quote; the quote still opens each exercise, the
heading is now what the exercise makes you do (ex #3 read your data). Every
cross-reference carries the name, so 'Ex #7' never appears bare.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
Also turns the exercise selector into an explicit if/then table. 7 and 8 were
bundled under 'about to report a result'; they now have their own conditions.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
Two different rules share the name fail fast. Nanda's is killing a doomed
direction early; this one is crashing on the error instead of carrying on.
Kept apart on purpose.
The order alone is unfollowable for an agent that has genuinely run dry.
Left out the 5-minute-experiment arithmetic, which gives the wrong number
for hour-long architecture runs.
His text, spelling fixed and voice kept, plus the three quotes from the cache
that back it: Bekman flagging his own overloaded heading, the tuning playbook
on two things sharing the name learning_rate, and Lones on which AUC.
gwern's own page concludes the tank story did not happen, so citing it
undercut the exercise. Zech is peer-reviewed with the in-site against
out-of-site AUC pair; the fastbook case covers the tabular version.
Comment review mode only, no prose changed. Flags negative framing,
aphoristic closers, and three places where the rewrite made wassname's
hedged claims stronger than his original message.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
Textbook order: collected advice, then his comment on how it applies to
LLMs, then the exercises. Content left for him to write.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
Small is under a paragraph; large means work like comparing against a
reference repo. Do all applicable small ones, pick one large one.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
Quotes come from the docs/evidence cache and were previously unused. Reuses the
existing footnote style, 14 new keys. Carries the source doc's coverage warning:
mode 6 has only three quotes and mode 7 has none that name similarity probes.
Source is his own message of 2026-08-25, spelling fixed and slightly more polite as he
asked, with each mistake pointing at the exercise that answers it. Also adds his rule
that a job is never abandoned without doing the exercises, one at a time.
Two gaps the existing 13 did not cover, found by mining the evidence cache against
wassname's list of common AI-agent failures. Quotes are verbatim from
docs/evidence/ (Steinhardt, Rahtz, Nanda, Goodfellow-Bengio-Courville).
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
Interpretations must read as evidence for or against a claim, never as a cause,
because a cause list gets picked from, called certain, and used to stop.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
Both were promised by the description and absent from the procedure. P2 now
prints the formatted examples; P3 opens with the bug assumption.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
A sonnet subagent had ml-debug listed, with 'after a run finishes or crashes' in
the description, and did not invoke it for 'read the last pueue log'.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
Sonnet routed to P1 correctly then declined it: 'the task only asked to read, not
to audit or act'. Reading a log is exactly when the measurement table gets filled.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
A sonnet subagent read the skill, summarised the ritual list, ran none of them,
and closed with 'everything checks out'. Phase triggers do not fire; a sentence
the agent watches itself write might.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
ml-bench measured standalone answers, not whether an agent in a loop follows
principles. Those are different claims and the first does not support the second.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
The run index is a cache key, not a seed, so subtracting run 3 from run 3 is
arbitrary. Unpaired the difference is +0.023 +- 0.044; by question +0.023 +- 0.031.
Still not pushed.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
+0.135 was one answer per arm, and it is draw 1 of three. Draws 2 and 3 read
-0.007 and -0.060, so the mean is +0.023 with sd 0.102. Not pushed: wassname
should read this before it goes public.
Co-Authored-By: Claudypoo <288921227+claudypoo@users.noreply.github.com>
