tessl/code-review-lens-creator
Author a review lens for Tessl Code Review: the reviewer skill that decides what `tessl code review` and the Tessl Code Review Action look for in a diff. Frames the review question and the bar a finding has to clear, drafts the lens, validates it by running it against changes that should and should not trip it, backtests it against pull requests that already carry review feedback, then packages it as a pinned registry ref a workflow can select. Use when someone wants their code review to catch something it misses, to stop flagging something it should not, or to write, fork, tune, or debug a code review lens. Not for `tessl review` rubrics, which score skill quality rather than code.
90
97%
Does it follow best practices?
Impact
96%
1.29xAverage score across 2 eval scenarios
Passed
No findings from the security scan
task.mdevals/scenario-2/
Make our code review catch the thing we keep catching by hand
Problem Description
We run Tessl Code Review on the payments service with the four default lenses. It is useful, but the same class of comment keeps coming from human reviewers instead, and it is the class that costs us most when it slips through: changes that behave correctly but cannot be diagnosed once they are running.
pr-feedback.md collects the review comments from last quarter. Not everything in there is the concern we mean.
We do not run this lens yet. It will be a fifth lens alongside the four defaults, and we are not turning it on until we have reason to trust it.
Output Specification
Produce two things in the workspace:
review-lenses/review-observability/SKILL.md, the lens itself, ready to run as a reviewer skill.validation-plan.md, saying how we should check it before we adopt it.