agentsclimarketplace

Auto tinker learn

Skill NolanCassidy/auto-tinker/skills/auto-tinker-learn

Local-first, chat-operated developer experiments, learning history, and approval-gated publishing.

Install
npx -y skills add NolanCassidy/auto-tinker --skill auto-tinker-learn

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 18 days oldThe repository was created 18 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Turn one experiment's evidence into a tailored tutorial, durable lesson, capability update, and next learning paths. Use after a run to explain what it proved, assess evidence-supported growth, or map prerequisites; use auto-tinker-review for broader status summaries.

SKILL.md

2.8 KB, 521 tokens by cl100k_base, as published. Nobody here has run it

Auto-Tinker Learn

Teach from evidence and the user's prior knowledge; do not inflate capability claims.

Invoke the CLI as auto-tinker; from an unlinked source checkout use npm --prefix <auto-tinker-product-repo> run cli -- followed by the same arguments.

Workflow

  1. Read the experiment, tests, decisions, failures, evidence snapshots, optional repository locations, source references, existing lessons, active main goal, and relevant graph nodes.
  2. Explain what the technology is, why it mattered to the goal, how the important parts work, what changed, what failed, and how verification supports the conclusion.
  3. Tailor detail to the profile and prior graph. Explicitly distinguish source facts, experiment observations, and inference.
  4. Persist the lesson with auto-tinker lesson create --title <title> --summary <markdown>, linking --experiment, --capability, and --tags.
  5. Append the candid narrative with auto-tinker journal append --kind private-journal when the run has a meaningful outcome. Keep readme, changelog, and public-story records distinct and linked.
  6. Run auto-tinker index and auto-tinker graph, then verify the experiment, lesson, capabilities, technologies, and next candidates are linked.
  7. Propose multiple next branches when useful; enqueue only those the user accepts.

Read lesson-and-capabilities.md for the lesson shape and evidence stages.

Rules

  • Failed and rejected hypotheses are valuable lessons when documented honestly.
  • Do not advance a capability from exposure to mastery because code was generated or one test passed.
  • Cite exact experiment/test/source record IDs in durable lesson content.
  • Preserve lessons and capability nodes when repository locations are missing or unverified; report that limitation instead of dropping the knowledge or inventing a URL.
  • Explain whether the result advanced the active main goal, informed a supporting goal, or proved a distraction was not worth continuing.
  • Keep publication-oriented storytelling separate from the full private lesson.
  • Preserve the user's writing voice only when profile evidence supports it; otherwise draft plainly for review.
  • Default narrative is first person and concrete: why I picked it, what interested or annoyed me, attempts/struggles, what I changed, how I verified it, and what I learned. Keep private detail out of public-story output.

Return the lesson path/ID, capability changes with evidence, graph links, unresolved questions, and ranked next learning branches.

What ships with it: 2 files

2.6 KB alongside SKILL.md

agents/

references/

Gives 0 of the 12 instructions most learn study skills give in 521 tokens

Counted across 546 of the 573 authors here whose files we hold, read 2026-08-07

  • Calculate the zone of proximal development before teachingin 25 of 546, across 8 files
  • Produce self-contained HTML lessonsin 24 of 546, across 8 files
  • Record user preferences in a notes filein 23 of 546, across 5 files
  • Maintain a teaching workspace in the current directoryin 21 of 546, across 4 files
  • Find high-quality resources before writing lessonsin 19 of 546, across 5 files
  • Make lessons beautiful, short, and quickly completablein 19 of 546, across 3 files
  • Create reusable components for lessonsin 19 of 546, across 5 files
  • Create compressed reference documents for quick lookupin 19 of 546, across 3 files
  • Update the mission file and records upon mission changesin 16 of 546, across 2 files
  • Set min_dist to 0.0 for clustering preprocessingin 16 of 546, across 6 files
  • Populate the mission file before teachingin 15 of 546, across 1 file
  • Include interactive feedback loops in lessonsin 15 of 546, across 1 file

Said here and by no other author read

  • teach from evidence and prior knowledge
  • read experiments, tests, evidence, and existing lessons
  • explain what changed, failed, and how verification supports conclusions
  • distinguish source facts, observations, and inference explicitly
  • persist lessons linking experiments, capabilities, and tags
  • append candid narrative to private journal for meaningful outcomes

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,984. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.