Research

"Evaluate DreamerV2 runs from JSONL and TensorBoard artifacts,

"Use when adding or debugging classifier/collision guidance, a

"Use the DeepXiv Reader for hosted arXiv/web agentic research,

"Use the DeepXiv SDK for citation-aware academic and web research

"Route DeepResearch/WebAgent family tasks to the right variant,

"Route potential-target and publication notes for the coronavirus

"Optional ColBERT Baleen multi-hop retrieval extension: HopSearcher

"Route multi-label classification and regression label-issue

"Routes unstable cleanlab experimental helpers for low-memory

"Validate ClawBio runs, benchmarks, reproducibility records, action

"Interpret Chinese-BERT-wwm benchmark data sources, archive

"Validate a BindCraft target PDB and construct a target JSON with

"Inspect BindCraft output folders and CSVs, explain filter outcomes

Preserve truthful experiment context with bounded memory, append-only history, safety signals, gates, and read-only progress inspection.

Launch, configure, steer, and safely stop a PROJECT_BRIEF.md-driven THINK→EXECUTE→REFLECT experiment loop.

"Choose ARIS research workflows, slash skills, skill groups, common

"Use ARIS, the Auto Research in Sleep skill-based research harness,

Analyze a local AI research paper, optionally using a local repository as supporting evidence, and produce a concise paper profile plus a validated module plan of at most five reusable skill modules.

Creates reusable Agent Skills from an AI research paper using the Paper2Skills Distiller workflow. Use in Creator mode when the user provides a paper PDF, paper URL, arXiv id, paper title, or paper/repo pair, or asks to convert a scientific paper into skills with recovery experiments. Prefer a TOML run config; ask for missing paper, repo, recovery, runtime, and budget fields before expensive work.

Create Jira tickets from incident RCA findings. Automatically generates structured tickets with root cause analysis, severity classification, and remediation steps.

Use when performing chaos engineering experiment — chaos engineering experiment template covering hypothesis definition, blast radius containment, experiment execution, observation, and learning. Supports Chaos Monkey, Litmus, Gremlin, and manual fault injection. Use for resilience validation and failure mode discovery.

Use when working with Statsig — statsig feature gate management, dynamic configs, experiment analysis, and metric tracking. Covers gate configuration, rule-based targeting, holdout groups, A/B test results, pulse metrics, and layer management. Use when managing feature gates, reviewing experiment results, analyzing metric lifts, or configuring dynamic configs in Statsig.

Use when working with Sourcegraph — sourcegraph code intelligence platform management covering repository indexing status, code search capabilities, batch change tracking, code insights monitoring, precise code intelligence status, and user access auditing. Use when investigating search indexing issues, monitoring batch change progress, reviewing code intelligence coverage, or auditing instance configuration.

Use when working with Posthog — postHog analytics management — monitor events, feature flags, experiments, session recordings, and ingestion health. Use when reviewing event volumes, managing feature flags, debugging tracking, or inspecting experiment results.