Rycen7822

Use DeepScientist from Codex CLI through the native dsctl adapter for research semantics/provenance: quests, memory, artifacts, experiments, strict literature workflow, paper/resource operations, and formal ds_bash_exec evidence. Routine file/search/edit,…

Use when a quest needs one or more follow-up runs such as ablations, robustness checks, error analysis, or failure analysis after a main experiment.

Use when a quest needs to attach, import, reproduce, repair, verify, compare, or publish a baseline and its metrics.

Use when the quest needs an explicit go, stop, branch, reuse-baseline, write, finalize, reset, or user-decision transition with reasons and evidence.

Use when a quest is ready for a concrete implementation pass or a main experiment run tied to a selected idea and an accepted baseline.

Use when the quest is ready to consolidate final claims, limitations, recommendations, summary state, and graph exports before stopping or archiving.

Use when a quest does not start from a blank state and the agent must first audit, trust-rank, and reconcile existing baselines, results, drafts, or review materials before choosing the next anchor.

Use when an algorithm-first quest should manage candidate briefs, optimization frontier, branch promotion, or fusion-aware search instead of the paper-oriented default loop.

Write CodexScientist-aware implementation, experiment-roadmap, code-only, and formal-command plans with TDD and durable quest outputs.

Use when a quest needs one or more follow-up runs such as ablations, robustness checks, error analysis, or failure analysis after a main experiment.

Use when a quest needs to attach, import, reproduce, repair, verify, compare, or publish a baseline and its metrics.

Use when the quest needs an explicit go, stop, branch, reuse-baseline, write, finalize, reset, or user-decision transition with reasons and evidence.

Execute CodexScientist project experiment command documents with durable cs_* logging, manifest validation, and cautious baseline-gate handling.

Use when a quest is ready for a concrete implementation pass or a main experiment run tied to a selected idea and an accepted baseline.

Use when the quest is ready to consolidate final claims, limitations, recommendations, summary state, and graph exports before stopping or archiving.

Use when a quest needs concrete hypotheses, limitation analysis, candidate directions, or a selected idea relative to the active baseline.

Delegate coding tasks to the OpenAI Codex CLI agent for implementation, refactoring, PR review, batch issue fixing, and autonomous repository work.

Configure, troubleshoot, and extend Hermes Agent itself: CLI setup, config, tools, skills, gateway, providers, plugins, MCP, cron, memory, and agent development.

Codex built-in skill for structural code search with ast-grep when queries depend on AST shapes, call patterns, decorators, scopes, or required child nodes.

Codex curated GitHub skill for general GitHub app workflows: repository metadata, issues, pull requests, comments, and connector-backed context.

Codex curated Hugging Face skill for preparing, running, and packaging community evaluation tasks and benchmark submissions.

Codex curated GitHub skill for addressing pull request review comments, resolving threads, applying requested changes, and reporting verification.

Codex curated GitHub skill for debugging and fixing failing GitHub Actions checks on pull requests with gh logs and focused patches.

Codex curated Hugging Face skill for building, debugging, and deploying Gradio demos and Spaces interfaces.

Review changes or review feedback for actionable defects and realistic impact.

Simplify selected code for clarity and maintainability while preserving intended behavior.

Explain how code works or why a design exists using relevant source and project history.

Diagnose bugs, failed checks, regressions, or performance problems with an unknown cause.

Write or revise long, source-grounded documents that need staged drafting or recovery.

Verify requested behavior through a running UI, CLI, service, or installed plugin.

Assess skill effectiveness from relevant agent history or scoped execution evidence.

Resolve consequential requirements, module, data-model, API, or migration design choices.

Operate Ponder-Forge complex-problem workflows through the pure CLI.

Thin workflow router for Idea-Spark shared-ledger review rooms: parent-orchestrated r1-r4 discussion-until-gate, subagent role contracts, CLI-first operation, and standalone handoff reports.

run nvmini for shortest NVIDIA GPU checks; fall back if needed.

run nvclean instead of nvidia-smi; fall back if needed.

Use proactively when complex or iterative work may produce durable HY Memory/Hermes memory: recall, save, verify, clean, or migrate reusable preferences, workflows, debugging lessons, and tool/API quirks without saving noisy task logs.

In large or many-file projects, prefer cxs over raw rg/grep to reduce context waste.

Use proactively when complex or iterative work may produce durable EverOS/Hermes memory: recall, save, verify, clean, compress, or migrate reusable workflows, debugging lessons, tool/API quirks, and agent cases without saving noisy task logs.

Install, verify, and operate the obclip CLI to clip live web pages into Markdown or Obsidian notes. Use when Codex needs to install the @harris7/obclip npm package, choose between default Playwright Chromium and a real Chromium executable/profile, run…

Schedule Codex-native same-thread compaction at a settled phase boundary, inject bounded continuity state, and continue automatically. Use only after active commands, tests, approvals, edits, and subagents have finished.

Delegate an unresolved repository fact or relationship that would otherwise require substantial new reading. Keep local reads, design, edits, and tests in the main agent.