keellisted

Measure whether a codebase's verification actually touches the world, or whether it is checking itself. Keel gathers the verification edges it can read in a target (GitHub Actions, CircleCI, GitLab CI, Travis, package scripts, Makefile, Rakefile, pyproject and a dozen tool configs) — and reports the surfaces it recognised but could not parse, so blindness never passes for absence. It classifies each edge as anchored, self-referential, or unknown by asking whether the actor being verified can write to the signal's producer, and reports a grounding ratio. Novel cases are judged by the agent and then crystallized into probes — small reviewable scripts — so repeat shapes get cheaper every run. Then routes each ungrounded check to an anchored signal that already exists in the same graph. Use when: (1) auditing whether AI-generated or agent-maintained work is genuinely verified, (2) assessing how AI-native a codebase actually is — measured by what fraction of its verification the agents cannot author, (3) reviewing
broomva/skills · ★ 3 · AI & Automation · score 75

Install: claude install-skill broomva/skills

# Keel A ship's keel is the reference line everything else is measured from. It is also what keeps the ship from capsizing — an even keel is a stability property, not a decoration. Keel measures one thing: > **A check is only a check if the signal it reads comes from somewhere the > thing being checked cannot write to.** Most verification in an agent-maintained codebase fails this. An LLM reviews what an LLM wrote. A doc is validated against another doc. A status field says "passed" because something set it to "passed". The pipeline is green and nothing has been verified. ## Where the predicate comes from Silver & Sutton (*Welcome to the Era of Experience*, 2025) argue that agents must replace human-prejudged reward with **grounded** signals, "that arise from the environment itself", because a human rater imposes "an impenetrable ceiling on the agent's performance" and, ungrounded, an agent "will become an echo chamber of existing human knowledge." Grounded and independent are not the same property, and the paper only asks for the first. Its own signal inventory — "cost, error rates … profit, sales … visits, yields, stocks, likes … accuracy" — mixes producers the actor cannot touch with producers it operates; and the success case it cites, **execution feedback** ("the agent actually runs code and observes what happens"), is maximally grounded *and* the canonical reward-hack surface, because the agent writes the test. Keel asks the second question — *and can the thing