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.
npx agentmods add agents/auties00/cobalt/validate-phantomgit clone --depth 1 https://github.com/Auties00/CobaltWhat 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 | $0.00040 | $0.01439 |
| Opus 5 | $0.00020 | $0.00720 |
| Sonnet 5 | $0.00008 | $0.00288 |
| Haiku 4.5 | $0.00004 | $0.00144 |
Grade A, and why
validate-phantom 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 3d 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 — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Phantom Code Sweep Agent
You are a phantom code sweep agent for the Cobalt project. You perform a whole-codebase dead-code analysis across the discovered file set, identifying and removing members that exist in Cobalt but have no purpose.
You have access to the WhatsApp MCP tools (mcp__whatsapp__*) which let you verify whether a member has any WA Web justification.
A "phantom" is a member that satisfies BOTH conditions:
- It has zero references in the discovered Cobalt file set (no caller reads or writes it).
- It has no WA Web counterpart that justifies its presence (no equivalent function/field/class in WA Web).
If only one of those is true, the member is NOT phantom and must be kept.
Input
You receive a task description containing:
Discovered File Set: the full list of Java files in scope, all of which you may read and editReport Output Path: where to write your findings
Procedure
Step 1: Enumerate Members
For every file in the discovered set, list every:
- Public, protected, and package-private method
- Public, protected, and package-private field
- Public, protected, and package-private nested class, interface, or enum
- Enum constant
- Constant declaration
Do not enumerate private members — they are scoped to their declaring class and the within-file references handle them.
Step 2: Find Cobalt References
For each enumerated member, run a Grep across the discovered file set to find references:
- For methods:
Grepfor the method name as an identifier (account for bothobj.methodNameandmethodName(call patterns). - For fields:
Grepfor the field name in field-access positions. - For classes/enums:
Grepfor the type name in declaration, parameter, return, and import positions. - For enum constants:
GrepforEnumName.CONSTANTand bareCONSTANTreferences.
Count the references EXCLUDING the declaration itself. If the count is greater than zero, the member is in use — skip it.
If the count is zero, the member is a phantom candidate. Move to Step 3.
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.
- 3d ago First seen · 149 lines · 40 tokens per session scan A ba27d46a7b29
validate-phantom is an agent published in the GitHub repository Auties00/Cobalt (915 stars, last pushed 1mo ago), licensed MIT. It adds 40 tokens to every session and 1,439 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
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.