---

name: autoresearch
description: "Autonomous iteration loop: modify, verify, keep/discard against any metric"
version: 2.1.4

Autoresearch — Autonomous Goal-directed Iteration

Safety Invariants (all subcommands)

• Never push, publish, or deploy without explicit user approval.
• Bounded by default. Override with Iterations: unlimited.
• All results logged to autoresearch/{subcommand}-{YYMMDD}-{HHMM}/ directory.
• Chain handoff via handoff.json. Evals reads *-results.tsv.

Subcommands

Command	Does	Default Iterations
	Iterate against a metric: modify → verify → keep/discard	25
plan	Convert a goal into validated Scope, Metric, Verify config	N/A
debug	Hunt bugs: hypothesize → test → falsify → repeat	15
fix	Crush errors one-by-one until zero remain	20
security	STRIDE + OWASP audit with red-team personas	15
ship	Ship through 8 phases: checklist → dry-run → deploy → verify	N/A
scenario	Generate edge cases across 12 dimensions	20
predict	5 expert personas debate before implementation	N/A
learn	Scout codebase → generate docs or wiki → validate → fix loop	10
reason	Adversarial debate with blind judges until convergence	8
probe	8 personas interrogate requirements until saturation	15
improve	Research ICP challenges, discover improvements, generate PRDs	15
evals	Analyze iteration results: trends, plateaus, regressions	N/A
regression	Regression stability gate: baseline vs candidate, verdict STABLE/UNSTABLE	N/A

Universal Flags

Flag	Applies To	Purpose
Iterations: N	All looping	Set iteration count
Iterations: unlimited	All looping	Opt-in unbounded
--evals	All looping	Mid-loop checkpoints + final summary
--evals-interval N	All looping	Override checkpoint frequency
--chain <targets>	All	Sequential handoff after completion
--<subcommand>	All	Shorthand for --chain <subcommand>