agent-config

Slow, exhaustive planning sessions: question the user one decision at a time until requirements are fully confirmed, then optionally chain into research scouts, a readiness audit, an architecture draft, and a staged roadmap. Trigger on "help me plan", "let's design", "I want to build", "deep planner", or /plan. Built-in plan mode is the lightweight alternative; this is the opt-in careful path.

Answer a how-to / syntax / config / factual question from a SOURCE the user points at (a docs URL, or a local file/directory of docs), not from memory, and cite the exact location so they can read around it. Trigger on a doc link or local path plus a question, or tool/software syntax where a stale answer would be wrong (NWChem, Slurm, a library's API). For community sentiment use recent-research; for a literature survey use deep-research.

Audit a repo's docs against its code and bring them back in sync: inventory all docs, fan out doc-auditor sub-agents to verify every concrete claim, report drift with proving paths, then fix and rebuild on your call. Trigger on "check the docs", "are my docs current", "audit the documentation", "docs drift", or a pre-release doc pass. NOT for answering questions from docs (doc-grounded) or prose quality (writing reviewers).

Pressure-test a real decision through five advisor lenses, anonymous peer review, and a single synthesized verdict (Karpathy's LLM Council on sub-agents). Trigger on 'council this', 'run the council', 'pressure-test this', 'stress-test this', 'debate this', or a genuine 'should I X or Y' with real stakes and options. Not for factual lookups, yes/no questions, or creation tasks. Claude Code only.

Scanner-grounded security audit, distinct from the built-in /security-review: runs the right scanners (semgrep, osv-scanner, trivy, gitleaks, bandit), queries live CVEs via MCP, walks a 24-category threat-model checklist, and produces a triaged report. Trigger on "deep security review", "full security audit", "check for CVEs", "scan my dependencies", or proactively after changes touching auth, crypto, deserialization, or dependency upgrades. Routine file/network I/O does NOT trigger; offer it in a sentence instead.

Macro-first pipeline for multi-page documents where structure matters as much as prose (proposals, papers, grants, reports): intake -> outline contract -> draft -> developmental, structural, specificity, voice, and persona reviews -> human-writer copy pass. Trigger on "deep writing pass", "this reads like AI even with no AI tells", or any long-form piece for a specialist audience. Under ~1 page use human-writer; for critique only, editor.

Use any time you generate or rewrite prose for the user — emails, documents, posts, proposals, any written content — so it reads like a sharp human wrote it: direct, specific, no AI tells. Also trigger on "make this sound less like AI", "rewrite this", "tighten this", or "clean this up" said of prose. NOT for code: "clean this up" pointed at code means refactoring, not writing. For critique without rewriting, use editor.

In-depth, tool-grounded code review, distinct from the built-in /code-review. Default scope is the WHOLE codebase (runs linters/tests, reads architecture, checks for AI slop, renders a Continue/Refactor/Rebuild verdict); scopes down to a deep change review only when the user names a PR, commit, or diff. Trigger on "deep code review", "review this codebase", "is this good code", "audit this project". For "is this exploitable?" use security-review-deep.

Design slide content and narrative for decks, talks, and pitches — what each slide says and why, not visual layout. Trigger on "help me with my deck", "what should this slide say", "plan my talk", "build out my presentation", or an outline the user wants turned into slides. Substantive content, not corporate filler; knows the user's presentation-template repo.

Check current community and web sources (Reddit, HN, forums, news) before answering on fast-moving topics. Trigger on what's new/best/trending, and whenever recency matters even unasked: version numbers, prices, release status, "still", "latest", the current state of a named tool or model. Skip when a stale answer wouldn't mislead. For academic literature use deep-research.

Generate a self-contained Markdown handoff document — plus a file inventory and proposed workspace layout if the session produced files — for moving work to another AI agent. Trigger ONLY on explicit asks: "handoff doc", "session handoff", "I'm moving this to Claude Code", "wrap this up so I can continue elsewhere". Not for generic chat summaries.

Bootstrap a new project with my conventions so it never starts as a blank slate. Trigger on "new project", "start a repo", "scaffold this", "set up a project", or picking a name for something about to be built. Picks the matching template (Python science/ML, Rust CLI, TypeScript tool, MCP server), lays out the skeleton, and writes a project CLAUDE.md.

Diagnose a failed, stuck, slow, or pending Slurm job on Stampede3 (TACC). Use when the user reports a JobID that crashed, exited early, ran out of memory/time, "is stuck in queue", produced unexpected output, or was killed. Walks through sacct, output logs, partition state, and Stampede3-specific failure modes (ibrun misuse, scratch purge, AVX mismatch, SPR HBM OOM). To write a new job script, use stampede3-submit.

Build a Slurm sbatch script for Stampede3 (TACC). Use when the user on Stampede3 asks to "run X on the cluster", "submit a job", "make a batch script", or "request N nodes/GPUs". Covers serial, MPI, OpenMP, hybrid, and GPU (H100/PVC) workloads with the correct partition, ibrun launch, and module setup. For diagnosing a job that already failed, use stampede3-debug.

Disciplined debugging when the user reports a bug, error, crash, failing test, or wrong output. Trigger on "this is broken", "why doesn't this work", stack traces, or code that runs but gives the wrong result. Finds the real root cause instead of patching symptoms, avoids repeating failed fixes, and walks through the plan before applying it.

Format ALL responses to be dyslexia-friendly. Trigger when the user mentions dyslexia, reading difficulties, or asks for easier-to-read output ("hard to read", "simpler layout"). Once active, every response follows the formatting rules, even short replies.

Critique-style feedback on an existing draft (emails, essays, reports, proposals, docs): structured notes, not redrafted prose. Trigger on "review this", "give me feedback", "does this flow?", or a shared draft plus a request for thoughts. If the user wants the text rewritten instead of critiqued, use human-writer.

Use when reading or editing the user's open Microsoft Word, Excel, or PowerPoint documents on macOS through the office MCP tools (word_*, excel_*, ppt_*, run_applescript). Covers the read -> edit -> screenshot workflow, tool conventions (1-based indexes, [r,g,b] colors, sheet param), the composite tools to prefer, and what AppleScript can't do.