/agents / ml-researcher

ml-researcher

ML literature, benchmarks, reproducibility, and tooling-reuse research. Math-first + observable-classification discipline. Read-only. Use for any ML/RL question, paper review, sim/dataset selection, or before proposing a custom env / training loop.

tier	T2
category	hub
writes?	read-only

Agent({
  subagent_type: "ml-researcher",
  prompt: "<your task>",
  isolation: "worktree"
})

# on fork, .claude/agents/<id>/ contains:
spec.md       # prompt verbatim
plan.md       # phases + verify-criteria
progress.json # {ts,phase,pct,note} every 30s
chatlog.md    # rolling transcript
handoffs.md   # sub-agents + branches
review.md     # self-review on completion

# orchestrator at completion
$ kei-ledger validate <agent-id>
$ git merge --no-ff agent/ml-researcher-1714291200
$ git push forgejo HEAD

# or squash / reject / defer per AskUserQuestion

other T2 agents

[AGENT]T2

critic

Ruthless code critic finding anti-patterns, tech debt, security issues, bugs, and performance traps. Read-only gate — outputs severity-sorted findings with file:line evidence. No fixes, only reports.

[AGENT]T2

critic-anti-pattern

Detects code anti-patterns: god classes, deep inheritance, shotgun surgery, primitive obsession. Read-only.

[AGENT]T2

critic-bug

Detects bug patterns: off-by-one, error-swallowing, unchecked Result/Option, race conditions in shared state. Read-only.

◂ back to /agents