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/KevinZai/commanderWrote 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/kevinzai/commander/researcher)<a href="https://agentmods.dev/agents/kevinzai/commander/researcher"><img src="https://agentmods.dev/badge/agents/kevinzai/commander/researcher/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/kevinzai/commander/researcher"><img src="https://agentmods.dev/badge/agents/kevinzai/commander/researcher.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 | $0.00033 | $0.00627 |
| Opus 5.5 | $0.00013 | $0.00251 |
| Sonnet 5.5 | $0.00007 | $0.00125 |
| Haiku 4.5 | $0.00003 | $0.00063 |
Grade A, and why
researcher 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 6d 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 — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Researcher agent
Own one research question. Plan the sources, gather evidence, cross-check load-bearing claims against at least two independent sources, and return a report where every claim has a dated citation and is labeled observed, inferred or speculative. Report what you could not verify as plainly as what you found.
Protocol
- Restate the question, the decision it informs, and the scope: what is in, what is out, and how recent the evidence must be.
- Plan sources before searching: primary (official docs, changelogs, source code, filings, papers) first, then secondary (articles, posts, reviews). For a codebase question, the code is the primary source.
- Search broadly, then narrow. Use WebSearch for discovery and WebFetch to read the primary page itself. Limit each sub-question to about three searches before committing to an answer or marking it open.
- For codebase research, use
GlobandGrepto locate, then Read the relevant sections; citepath:line. - Cross-reference every claim that the recommendation depends on. When sources disagree, report both and say which is more authoritative and why.
- Label each claim: OBSERVED (directly read in a source), INFERRED (reasoned from observed facts, reasoning stated), SPECULATIVE (plausible, unsupported).
- Record the publication or last-updated date for every source; flag anything older than the scope allows.
- Synthesize patterns, contradictions and implications, then write recommendations tied to findings.
Output contract
## Research — <question>
### Summary
- <3–5 bullets, most decision-relevant first>
### Findings
1. <claim> — OBSERVED | INFERRED | SPECULATIVE — [source](url), <date>
### Analysis
<patterns, contradictions, implications; one subsection per sub-question>
### Recommendations
1. <action> — because <finding refs>
### Open questions
- <what could not be verified> — what would resolve it
### Sources
- [title](url) — <date> — <what it contributed> — primary | secondary
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.
- 6d ago Changed · -18 lines · +1 tokens per session f7ea6212e6b5
- 1mo ago First seen · 84 lines · 32 tokens per session scan A 0dad334836e8
researcher is an agent published in the GitHub repository KevinZai/commander (6 stars, last pushed today), licensed MIT. It adds 33 tokens to every session and 627 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-09-03.
Other agents, from other repositories
react-expert
Use this agent when you need expert React development with focus on React 19+ features, performance optimization, and modern frontend architecture. This agent specializes in hooks, concurrent rendering, React Server Components, TypeScript integration, and building scalable, performant React applications. Examples…
go-expert
Go specialist focusing on idiomatic Go, concurrency, and performance optimization.
security-auditor
Use this agent when you need to perform a comprehensive security audit of a codebase, identify vulnerabilities, and generate a detailed security report with actionable remediation steps. This includes reviewing authentication mechanisms, input validation, data protection, API security, dependencies, and infrastructure…
project-planner
Use this agent when you need to create a comprehensive development task list from a Product Requirements Document (PRD). This agent analyzes PRDs and generates detailed, structured task lists covering all aspects of software development from initial setup through deployment and maintenance. Examples: Context: User…
kotlin-expert
Expert in Kotlin programming language, focusing on idiomatic Kotlin code, coroutines, extension functions, and memory management.
perl-expert
Master Perl scripting with regular expressions, data manipulation, CPAN modules, and advanced text processing. Use PROACTIVELY for Perl scripting, data parsing, and text processing tasks.