report-assembler

A code-review agent combines results from several batches of file reviews into one report.

In plain words
What is it for?
It helps produce a CODE_REVIEW.md report, identify repeated code problems across files, and highlight tightly coupled changes or opportunities to share code.
Why use it?
It removes duplicate findings and orders issues by severity, making a large review easier to act on.

Agent

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/luongnv89/skills/report-assembler
Clone the repo
git clone --depth 1 https://github.com/luongnv89/skills
Per session 26 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 957 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 $0.00026 $0.00957
Opus 5 $0.00013 $0.00478
Sonnet 5 $0.00005 $0.00191
Haiku 4.5 $0.00003 $0.00096

Measured 2d ago against content hash 8037918e9b1e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

report-assembler 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 2d 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.

skills/code-review/agents/report-assembler.md · 163 lines

How it starts

The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Report Assembler Agent

Consolidate findings from parallel file-reviewer batches into a single CODE_REVIEW.md report with deduplication and severity ranking.

Input

Receives array of JSON outputs from file-reviewer:

{
  "batches": [
    { "batch_id": "batch-001", "findings": [...], "summary": {...} },
    { "batch_id": "batch-002", "findings": [...], "summary": {...} }
  ],
  "metadata": {
    "scope": "PR #123 | Full Audit",
    "base_ref": "main",
    "head_ref": "feature-x",
    "files_total": 50,
    "date": "2026-03-24"
  }
}

Process

  1. Merge all findings from batch results into a single list

  2. Deduplicate:

    • Group findings by (file, line, smell) tuple
    • Keep severity of highest priority duplicate
    • Merge descriptions, combine before/after examples
    • Preserve all unique fix suggestions
  3. Rank by severity:

    • Critical (top priority)
    • Major (should fix soon)
    • Minor (nice to have)
    • Info (optional suggestions)
  4. Cross-file analysis:

    • Identify duplicate code patterns across files
    • Flag shotgun surgery risks (one change, many files)
    • Note architectural coupling issues
    • Highlight opportunities for shared abstractions
  5. Categorize by type:

    • Security & Safety (critical focus)
    • Code Smells (bloaters, couplers, etc.)
    • Pragmatic Principles (DRY, orthogonality, etc.)
    • Maintainability
  6. Count by severity for summary table

Output Format

Generate CODE_REVIEW.md:

# Code Review Report

**Date**: YYYY-MM-DD
**Scope**: [PR #123 | Full Audit]
**Files Reviewed**: N
**Total Issues**: X

## Summary

| Severity | Count |
|----------|-------|
| Critical | X |
| Major    | X |
| Minor    | X |
| Info     | X |

## Critical Issues

### [Category]: Issue Title

**Files**: `path/to/file.ts:42` (+ X more occurrences)
**Smell**: [Code smell name]
**Pattern**: [If applicable - duplicate across files]

Description of the issue with context.

**Before**:
\`\`\`language
// problematic code
\`\`\`

**Suggested Fix**:
\`\`\`language
// improved code
\`\`\`

**References**: [links to applicable principles/patterns]

---

## Major Issues
[Sorted by file and line number]

...

## Minor Issues
...

## Info & Suggestions
...

## Recommendations

1. **Immediate actions** (Critical issues to fix before merge)
2. **Refactoring priorities** (Major issues for sprint planning)
3. **Architecture improvements** (Cross-file patterns, coupling, scalability)
4. **Team practices** (Broken windows, testing, documentation)

## Files Not Reviewed

[List of files skipped due to size/binary/excluded patterns]

## Methodology

- Code smells identified using catalog in `references/code-smells.md`
- Pragmatic Programmer principles from "The Pragmatic Programmer"
- Security checks cover OWASP Top 10 common patterns
- Maintainability scored on naming, complexity, coverage

---

*Generated by code-review skill v1.1.0 with subagent architecture*

Read the full file on GitHub · 163 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. 2d ago First seen · 163 lines · 26 tokens per session scan A 8037918e9b1e

Subscribe to this mod's changes

report-assembler is an agent published in the GitHub repository luongnv89/skills (121 stars, last pushed 2d ago), licensed MIT. It adds 26 tokens to every session and 957 once invoked, about $0.0001 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.