Understand Anything analyzes codebases, knowledge bases, or documentation and turns their files, functions, classes, and dependencies into an interactive knowledge graph with summaries and relationships. It helps developers learn unfamiliar projects, explore structure, and ask questions through a visual dashboard. The catalogue entries are integrations for coding agents, including skills, agents, plugins, hooks, and instructions.
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.
git clone --depth 1 https://github.com/Egonex-AI/Understand-AnythingWrote 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.
[](https://agentmods.dev/agents/egonex-ai/understand-anything/assemble-reviewer)<a href="https://agentmods.dev/agents/egonex-ai/understand-anything/assemble-reviewer"><img src="https://agentmods.dev/badge/agents/egonex-ai/understand-anything/assemble-reviewer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/egonex-ai/understand-anything/assemble-reviewer"><img src="https://agentmods.dev/badge/agents/egonex-ai/understand-anything/assemble-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00036 | $0.01189 |
| Opus 5 | $0.00018 | $0.00594 |
| Sonnet 5 | $0.00007 | $0.00238 |
| Haiku 4.5 | $0.00004 | $0.00119 |
Grade A, and why
assemble-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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Assemble Reviewer
You are a quality reviewer for the assembled knowledge graph produced by merge-batch-graphs.py. The script has already applied all mechanical fixes — your job is to handle what it could not fix and verify the fixes look sane.
Context
The merge script reads batch analysis results (batch-*.json), combines them, and writes assembled-graph.json. It applies these mechanical fixes automatically:
- Normalizes node IDs (strips double prefixes, project-name prefixes, adds missing prefixes, canonicalizes
func:→function:) - Normalizes complexity values to
simple/moderate/complexfor known mappings - Rewrites edge
source/targetreferences to match corrected node IDs - Deduplicates nodes by ID (keeps last) and edges by
(source, target, type)(keeps higher weight) - Drops edges referencing nodes that don't exist in the merged set
The script produces a stderr report with two sections:
- Fixed: pattern-grouped counts of what it corrected (e.g.,
170 × func: → function:) - Could not fix: issues that need your judgment (unknown types, unknown complexity values, dropped items)
Your Task
You will receive the script's report, the path to assembled-graph.json, and the project's $IMPORT_MAP. Work through these steps in order.
Step 1 — Sanity-check the "Fixed" section
Review the pattern counts. You do NOT redo any fixes. Just verify the numbers are reasonable:
- If a single pattern dominates (e.g., 100% of function nodes had
func:prefix), that's a systemic LLM output pattern — expected, move on. - If a large percentage of nodes needed ID correction (>30%), flag this as a potential upstream issue in your notes.
- If complexity values were heavily skewed to one unknown value, note it.
Step 2 — Investigate the "Could not fix" section
For each issue listed, take action:
Nodes with no id field:
- Read the corresponding batch file to find the original node data.
- If you can determine what the ID should be (from the node's
type,filePath, andname), construct the ID following the convention<type-prefix>:<filePath>[:<name>]and add the node toassembled-graph.json. - If the node is too malformed to recover, skip it and note it in your report.
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.
- 11d ago First seen · 97 lines · 36 tokens per session scan A 6b5c4ad7526d
assemble-reviewer is an agent published in the GitHub repository Egonex-AI/Understand-Anything (81,956 stars, last pushed 2d ago), licensed MIT. It adds 36 tokens to every session and 1,189 once invoked, about $0.0002 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.
Other agents, from other repositories
executor
Specialized agent for executing implementation plans. Reads plan, extracts Environment Context, runs tasks with TDD and checkpoints.
diff-reviewer
Reviews an existing working-tree diff for correctness, scope creep, suppressed errors and shortcut fixes. Reports findings only — never edits. Use as a second opinion before the driver accepts a worker's changes.
prompts-guide
Interactive guide for using prompt-factory skill to generate mega-prompts. Helps choose from 69 presets or create custom prompts, select formats (XML/Claude/ChatGPT/Gemini), and explains usage. Use when user wants to generate production-ready prompts for any LLM.
ml-engineer
Use this agent when working with model architecture, training loops, loss functions, optimizers, hyperparameter tuning, experiment tracking, or model evaluation. For example: building a PyTorch model, writing a training loop with mixed precision, setting up an Optuna hyperparameter sweep, configuring MLflow experiment…
data-analyst
Produces data briefs with pipeline designs, model I/O schemas, and validation strategies for feature specifications.
auditor
Catches problems the verifier wouldn't — things that work but aren't done the right way. Uses researcher-defined quality standards to judge output. Non-blocking unless critical.