Exec Subtle — Data & AI agent for Claude Code
FRONTIER-DO tier at max effort.
How to install Exec Subtle
Installs to ~/.claude/agents/lcs-adam-standing-orders-exec-subtle.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/LCS-Adam/standing-orders/HEAD/agents/exec-subtle.md -o ~/.claude/agents/lcs-adam-standing-orders-exec-subtle.md Restart Claude Code, or start a new session, for it to be picked up.
What Exec Subtle does
name: exec-subtle description: FRONTIER-DO tier at max effort. ULTRATHINK execution agent for the single subtlest workstream in a plan - derivation changes with mixed historical data, flags that must resist an existing self-heal, invariants with a crying-wolf failure mode. Writes code, tests, and commits in an isolated worktree. Reserve it; use exec-critical for ordinary correctness-critical work. tools: Read, Edit, Write, Bash, Grep, Glob model: {{TIER_FRONTIER_DO}} effort: max reasoning: m
Alternatives in Data & AI
- Vision — GSD-2 is the orchestration layer between you and AI coding agents 5.7k ★
- Council Sutskever — Council member 4.2k ★
- Stack Auditor — Audits a codebase's AI agent stack against the live best-of-Agent-Harnesses dataset — finds which harnesses th 801 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Omn Tech Lead.Agent
Tech Lead of the AI Engineering Framework. Owns delivery feasibility, sequencing, and risk-controlled executio
Benchmark Specialist
Use for benchmark execution, monitoring, and data collection during evaluation runs. Invoked for running tier
Executor Heavy
Orchestra heavy executor (Opus). Use for the hard tier of execution work orders — algorithmically hard cores (
Task Reviewer
Strict quality reviewer for implementation task files. Validates completeness, self-containment, and clarity f
Constitutional Reviewer
Multi-pass self-review that forces write-critique-rewrite-verify cycle. Use on critical code paths — auth, bil
Refiner
Multi-pass self-review that forces write-critique-rewrite-verify cycle. Use on critical code paths — auth, bil
Related Skills
Massu Estimate
Effort estimation with complexity scoring, codebase impact analysis, and historical comparison
Codex Consult
Consult OpenAI Codex (codex exec) for a second opinion on a prompt. Handles the argv-hang, stdin-mode, reasoni
Vet Skill
Static pre-install scanner for AI agent skill bundles — flags prompt-injection phrasing and exfil/remote-exec-