Repository navigation
Record session-environment lessons and separate review axes - #8
Merged
Merged
Conversation
Fit navigation, tool economy, and no-op steering lessons into the existing correction ladder, and ask review to keep plan or spec misses separate from repository standards or simplicity. Co-authored-by: Dennis Geldmacher <geldmacher.dennis@outlook.de>
cursor
Bot
force-pushed
the
dg-cursor/session-lessons-review-axes-c2eb
branch
from
October 7, 2026 07:32
dbcc711 to
314713e
Compare
geldmacher
marked this pull request as ready for review
October 7, 2026 07:34
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Context
This follows a review of Matt Pocock's skills v1.3.1 (
mattpocock/skillsat24fe0ef7737efae15c87225755e9f6f5965e4888), announced in https://x.com/mattpocockuk/status/2107062474914550098. The sources read in detail were/retroandcode-review.Rebased onto current
mainafter v7.5–v7.7 and the vague-scope planning note from #9. That section stays. The agent-environment note is added beside it, and the grill-me / glossary reject is not repeated.Two gaps were worth taking. The rest was left out on purpose:
/implement-spec, and the ticket-graph implementer flow around it, conflicts with Auto-Work and the written plan handoff. Planning stays the high-effort step. Implementation follows that plan. Host configuration chooses models; this plugin does not hardcode them.GLOSSARY.mdand grill-me / grill-with-docs would add a required glossary and an interview. That fights the lean handoff, and a clear assignment already needs no interview. Vague-scope planning already records that reject./retrocommand, no requiredCODING_STANDARDS.md, no parallel review subagents, and no Fowler smell list. Automated checks were already on the correction ladder from Prefer proportionate prevention when corrections recur #7.What changed
references/learning-work.md): the session or its log is a source for lessons about the agent's environment. The added cases are navigation, wasteful tool use, and steering instructions that do not change behavior. A mechanical lesson becomes a check. Save a judgment lesson as a reviewer standard outside implementation context when Review confirms it is reusable and learning is commissioned; until then it stays a review finding or an open candidate. The existing order (structure, then CI, then bounded guidance, then human review) stays. A few nearby sentences were tightened so the addition fits the frozen context limits.skills/review-work/SKILL.md): findings stay on two axes in the same read-only pass: deviation from the plan or spec, and repository standards or simplicity. On a simplicity question, use Efficiency's guidance when that plugin is already available.docs/methodology-sources.md, with an Unreleased changelog note. No version bump.Walkthrough rows in
docs/behavior-validation.md, plus short notes in the working guide and project-improvement page, match those decisions.Why
The correction ladder already preferred a structural fix or a check over another prose rule. What it did not say is to read the session itself for how the agent moves through the repository, or to keep judgment standards on the reviewer so implementation context stays lean. Splitting review findings the same way stops a standards pass from hiding a missed requirement, and the other way around, without a second reviewer.