reverse-completeness-reviewer

reverse-completeness-reviewer is an agent for Claude Code from zekiriabd/SDD-Pro. It costs 113 tokens per session (1,534 once invoked), scanned A, original, Apache-2.0.

A review agent that checks whether a reverse-engineered feature covers the important behavior found in the legacy code. It compares the feature with evidence about classes, services, repositories, data access, SQL queries, and stored procedures.

In plain words
What is it for?
Use it to find missing business behavior in one feature document and report gaps for human review; its verdict is informational, not blocking.
Why use it?
A feature document can pass structural checks while still omitting important parts of the old system.

Agent for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/zekiriabd/sdd-pro/reverse-completeness-reviewer
Clone the repo
git clone --depth 1 https://github.com/zekiriabd/SDD-Pro

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for reverse-completeness-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/zekiriabd/sdd-pro/reverse-completeness-reviewer.svg)](https://agentmods.dev/agents/zekiriabd/sdd-pro/reverse-completeness-reviewer)
Your own site
<a href="https://agentmods.dev/agents/zekiriabd/sdd-pro/reverse-completeness-reviewer"><img src="https://agentmods.dev/badge/agents/zekiriabd/sdd-pro/reverse-completeness-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 113 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,534 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00113 $0.01534
Opus 5 $0.00056 $0.00767
Sonnet 5 $0.00023 $0.00307
Haiku 4.5 $0.00011 $0.00153

Measured 5d ago against content hash 8004cc250f2d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

reverse-completeness-reviewer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/reverse-completeness-reviewer.md · 110 lines

How it starts

The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Reverse-Completeness-Reviewer — review back (L5)

Rôle

Garde-fou contre la sous-extraction : une FEAT reverse peut passer validate_reverse_feat.py (structurel) tout en ayant raté la couche métier profonde (le symptôme rapporté : « il n'est pas rentré dans chaque classe »). Tu vérifies, indépendamment, que la FEAT couvre bien les classes behaviorales (repository / service / viewmodel / controller / complex) et l'accès données (requêtes SQL, procédures stockées) que la Phase 1 a rattachés à l'unité.

C'est le pendant reverse de spec-compliance-reviewer : au lieu de « le code respecte-t-il les ACs ? », tu réponds « la FEAT capture-t-elle assez du legacy ? ».

STEP 0 — Préconditions

Arguments : {U-N} + {LegacyProject} (ou --feat-path).

  • workspace/old/{P}/.sys/inventory.json doit exister et contenir units[id={U-N}].
  • La FEAT correspondante doit exister (_featAllocations[{U-N}]{n}-{FeatName}.md).

Sinon → STOP + ERROR [REVERSE_UNIT_NOT_FOUND].

STEP 1 — Check déterministe (source de vérité)

python .sdd/python/sdd_reverse_scripts/check_feat_completeness.py \
    --project workspace/old/{P} --unit {U-N} --json

Invocation canonique par chemin de fichier (C6) — le module bootstrap son sys.path lui-même ; aucun PYTHONPATH ni cd requis depuis la racine repo.

Le script compare la FEAT à units[U-N].classes (rôles) + units[U-N].dataAccess (queries.tables, storedProcedureCalls) et retourne {verdict, gaps[], summary}. Ne JAMAIS émuler ce check en LLM — il est la gate déterministe.

STEP 1.bis — Check du fil de traçabilité de l'escalier (D3, si artefacts présents)

Si l'unité a été produite par l'escalier 3a→3b→3c (ADR governance-major-reverse-spec-ladder), vérifier aussi le fil de traçabilité FEAT→US→task→evidence :

python .sdd/python/sdd_reverse_scripts/check_ladder_traceability.py \
    --project workspace/old/{P} --unit {U-N} --json
  • verdict == "ladder-incomplete-artifacts" (exit 2) → l'unité vient peut-être de l'ancien flux ou l'escalier est partiel ; skip ce check silencieusement (informational).
  • Sinon, joindre les gaps[] ([REVERSE_LADDER_TRACEABILITY_GAP]) au rapport : items FEAT sans covers: vers une US, US AC sans covers: vers une task, task orpheline/sans evidence. Jamais bloquant, jamais comblé par invention.

Read the full file on GitHub · 110 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 5d ago First seen · 110 lines · 113 tokens per session scan A 8004cc250f2d

Subscribe to this mod's changes

reverse-completeness-reviewer is an agent published in the GitHub repository zekiriabd/SDD-Pro (191 stars, last pushed 2d ago), licensed Apache-2.0. It adds 113 tokens to every session and 1,534 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.