old-coder vs supercov
Uncle Bob's strategy as a skill: the agent writes a SPEC you approve, then runs a gauntlet — tests, mutation, property-based, coverage, supply chain — and hands you an evidence report instead of code. — versus — Code quality and coverage for coding agents: a Rust CLI that scores files with Jev, runs your existing test command, and turns uncovered paths into small actionable targets.
Both refuse to let an agent call work done on vibes; old-coder enforces a SPEC-plus-gauntlet discipline as a skill, supercov enforces it with measured coverage and MC/DC obligations instead of process.
| old-coder | supercov | |
|---|---|---|
| Stars | 738 | 56 |
| Forks | 56 | 1 |
| Language | Python | Rust |
| License | MIT | MIT |
| Last activity | 1 months ago | today |
| Topics | skills, coding | coding, evals |
| Curated connections | 5 | 4 |
old-coder — the curator's take
The right skill when you've accepted you won't read the diff: you read two documents instead — a SPEC of concrete must and must-not examples before any code, and an EVIDENCE report with real numbers from one fresh run you can reproduce with a single command. Effort scales with risk, so a typo fix runs two checks while anything touching money, auth or concurrency runs mutation testing, property-based tests and hostile inputs. It's plain Markdown, so any instruction-following agent can use it. Read the honesty clause carefully: the gauntlet turns your spec into executable evidence, it cannot prove the spec is complete or authenticate its own checkers — the demo's own report caught an unsound mutation runner behind an earlier green state.
supercov — the curator's take
Point your agent at it when the instruction 'write some tests' keeps producing tests that prove nothing: supercov runs the suite the repo already trusts, reports lines/branches/MC-DC, and lists the specific uncovered obligations so the agent has a target it can verify it hit. The quality half asks Jev yes/no questions per file (long_method, deep_nesting ...) and does the arithmetic locally, so every score decomposes into claims you can check against the source - about a cent per megabyte, cached by content. Coverage needs no account, key or reporter; quality needs a hosted TypeSafe key. Eight languages. Don't expect a verdict on correctness: coverage and the assertion map tell you what ran and what was asserted, not whether the behaviour is right, and the assertion score is agent-assessed by design.