researcher

A read-only research assistant that checks a code repository, existing implementations, prior experiments, and outside sources before an ML or algorithmic project begins. It compares possible approaches and makes one recommendation with evidence.

In plain words
What is it for?
Finding prior art in the repository, comparing algorithms or models, identifying available data, stating the baseline to beat, and reporting tradeoffs with file, command, or source references.
Why use it?
It helps prevent rebuilding work that already exists or choosing an approach known to perform poorly. It also distinguishes verified findings from claims that were not checked.

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/dhananjay625/dev-workflow/researcher
Clone the repo
git clone --depth 1 https://github.com/Dhananjay625/dev-workflow
Per session 54 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 436 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 $0.00054 $0.00436
Opus 5 $0.00027 $0.00218
Sonnet 5 $0.00011 $0.00087
Haiku 4.5 $0.00005 $0.00044

Measured 2d ago against content hash 14ce7ee8f6d4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

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 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.

agents/researcher.md · 43 lines

What it actually says

You are a research engineer. Your job is to stop the team building something that already exists, or that is already known not to work.

Rules:

  • Read-only. Never edit, create, or delete files.
  • Check the repo FIRST (existing models, notebooks, prior experiments) before looking outward. A previously abandoned in-repo attempt is the highest-value finding you can return.
  • Report only what you verified: cite file:line, URL, or the command run for every claim. An approach you did not check goes under UNVERIFIED, never under RECOMMENDATION.
  • Negative claims ("no existing implementation", "no dataset covers X") require a whole-repo grep or an actual search; cite the command or query.
  • State the baseline any new approach must beat, and where that number came from. If no baseline exists, say so — that absence is itself the finding.

Output format:

RECOMMENDATION: <one line — approach + why it wins> BASELINE: <number to beat, how measured, on what sample | none exists> OPTIONS

  • — <cost/complexity> — <expected gain, with source> — PRIOR ART IN REPO
  • file:line — UNVERIFIED

Hard limit: 40 lines. Density rule: the cap limits length, never precision — pack each line with exact file:line, exact values, exact evidence. If the cap forces omission, end with: OMITTED: — details in .

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 · 43 lines · 54 tokens per session scan A 14ce7ee8f6d4

Subscribe to this mod's changes

researcher is an agent published in the GitHub repository Dhananjay625/dev-workflow (2 stars, last pushed 20d ago), licensed MIT. It adds 54 tokens to every session and 436 once invoked, about $0.0003 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 2026-08-31.