Skip to content

feat(review): hunter-over-engineering — 7º track repo-wide de sobre-ingeniería - #4

Merged
ixjosemi merged 1 commit into
mainfrom
feat/review/hunter-over-engineering
Aug 10, 2026
Merged

ixjosemi merged 1 commit into
mainfrom
feat/review/hunter-over-engineering

Conversation

@dagoaie

@dagoaie dagoaie commented Jul 31, 2026

Copy link
Copy Markdown
Owner

feat(review): hunter-over-engineering — 7º track repo-wide de sobre-ingeniería

Adds a repo-wide over-engineering audit as the seventh specialty track of the hunter and hunter-max pipelines. Unlike the diff-scoped over-engineering-auditor (PR 2), this track scans the whole repository when prd.md names no narrower scope, hunts unearned abstraction / indirection / configurability / speculative generality / removable code, ranks findings by largest cut first, and reports the net lines and dependencies removable. Inspires from the Ponytail ponytail-audit skill, but is implemented natively: same hunter fan-out, consensus, statistics, and fingerprinting as the other six tracks.

This is PR 4 of a 4-PR program inspired by Ponytail — Convoy dogfooding itself through its own implement pipeline. Based on main; independent of PRs 1–3.

What changed

  • prompts/hunter-over-engineering.md (new): repo-wide over-engineering audit prompt in the standard hunter-* shape (specialty, scope, limitations, findings with severity/confidence/location/evidence/trigger/fingerprint), with the N1–N7/V1–V7 taxonomy and the seven tags (yagni, premature-abstraction, configurability, indirection, complexity, over-processing, scope-creep) defined inline so it is self-contained on main. Findings are ranked by biggest cut first; the report closes with the net lines and dependencies removable, and says so plainly when there is nothing to cut.
  • src/pipeline.ts: adds "over-engineering" to hunterTracks, registers the hunter-over-engineering agent (readOnly: true, temperature: 0.1), and wires it into the hunter fan-out as [fallbackModel, glmModel] — it flows into hunter-max automatically through hunterMaxTracks().
  • src/built-in-prompts.ts: registers the new prompt (sync test covers it).
  • prompts/hunter-report.md / prompts/hunter-max-report.md / README.md / src/config.ts: expected coverage updated from six to seven specialties (30 → 35 audits in hunter-max).
  • Tests: registry and hunter/hunter-max pipeline-shape asserts updated for the seventh track.

Verification

  • bun run typecheck — pass
  • bun test — 755 pass, 0 fail
  • bun run build — pass

Out of scope (follow-ups)

  • hunter/hunter-max specialty model roster unchanged for the existing six tracks; no new pipeline; review/refine/review-cc untouched.

- hunter-over-engineering work in progress (interrupted by model switch)
@dagoaie dagoaie self-assigned this Jul 31, 2026
@dagoaie

dagoaie commented Jul 31, 2026

Copy link
Copy Markdown
Owner Author

Hola, @ixjosemi:
te he invitado como colaborador (write) de dagocareer/convoy para poder añadirte como revisor oficial de este PR.

Cuando aceptes la invitación, te añado como requested reviewer.
Mientras tanto, tu revisión es bienvenida cuando quieras.

Un abrazo

@dagoaie

dagoaie commented Jul 31, 2026

Copy link
Copy Markdown
Owner Author

@ixjosemi te comparto esto por si te apetece echarle un vistazo, sin ningún compromiso — solo era para compartirlo 😄

Es un fork de Convoy en mi GitHub (dagocareer/convoy). Me pareció una herramienta muy útil para el día a día y le he añadido cuatro features que me faltaban. Lo curioso es que toda la implementación se ha hecho usando el propio Convoy (dogfooding): cada feature se escribió con su pipeline implement, con su PRD, sus auditorías y su revisión adversarial — Convoy revisándose a sí mismo. Y además, cada PR se fue apilando encima de la anterior.

Un resumen de cada una, en una frase:

  • PR feat(summary): diffstat scoreboard per fase en SUMMARY.md #1 — diffstat scoreboard: añade un marcador por fase en SUMMARY.md que muestra cuántas líneas/archivos toca cada paso del pipeline, para ver de un vistazo cuánto trabajo real hace cada fase.
  • PR feat(review): over-engineering auditor #2 — over-engineering auditor: Convoy ahora revisa también sobre-ingeniería en cada PR — abstracciones prematuras, flags que nadie usa, capas innecesarias — etiquetadas y con niveles, igual que ya revisaba código limpio, seguridad y bugs.
  • PR feat(review): debt-auditor #3 — debt-auditor: cuando una revisión difiere un hallazgo sin resolver, ya no se pierde: entra en un ledger de deuda técnica con un disparador para reabrirla; y si no hay forma honesta de justificar el aplazamiento, queda marcada como deuda «sin disparador».
  • PR feat(review): hunter-over-engineering — 7º track repo-wide de sobre-ingeniería #4 — hunter-over-engineering: Convoy revisa el repositorio entero — no solo lo que cambia en cada PR — buscando código inflado que podría recortarse, ordenado por cuánto se puede ahorrar en líneas y dependencias.

Están todas abiertas y mergeables por si te apetece revisarlas, pero vamos, sin compromiso.

@ixjosemi
ixjosemi merged commit f4aaeae into main Aug 10, 2026
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants