Agents Results — Development agent for Claude Code
agents_results.md.
How to install Agents Results
Installs to ~/.claude/agents/jgreathouse9-mlsynth-agents_results.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/jgreathouse9/mlsynth/HEAD/agents/agents_results.md -o ~/.claude/agents/jgreathouse9-mlsynth-agents_results.md Restart Claude Code, or start a new session, for it to be picked up.
What Agents Results does
agents_results.md
Agent-facing contract for **what an mlsynth estimator returns** and the **Definition of Done (DoD)** for migrating an estimator onto it. Read this together with `agents_estimators.md` (how an estimator is built), `agents_docs.md` (docs conventions), and `agents_tests.md` (testing layers).
**Status.** The contract is validated end-to-end on the three reference estimators **VanillaSC, FDID, TSSC** (see `mlsynth/tests/test_result_contract.py`). Roll it out one estimator
Alternatives in Development
- Leader — Central decision-maker that plans experiments and reflects on results 728 ★
- Theorist — Theoretical econometrician / mathematical statistician 250 ★
- Backlog MCP Validator — Validate the backlog FastMCP server against the CLI 64 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Hecras Results Analyst
Orchestrator Multi
Multi-agent parallel orchestrator using git worktrees — splits work across N agents, manages territory, merges
Scope Controls
Controls orchestrator — dispatches six subagents in two waves (org-wide issues, detections, dashboard ideas, p
Search Orchestrator
Multi-channel search orchestration — classifies complexity, fans out to channel subagents, deduplicates and ra
Serp Extract
Extracts and normalizes SERP snapshots with provider metadata, organic results, and SERP features.
Experiment Scientist
Analyze experiment results, respond to audits and reviews, update STATE.md, and plan the next experiment round
Related Skills
Context Mode — Benchmark Results
Benchmarked against real outputs from popular Claude Code MCP servers, Skills, and dev tools. All fixtures cap
Axiom — Development Changelog
Purpose: Historical record of Axiom development milestones, TDD testing results, and architectural decisions.
Swarm Collect
Collect and synthesize results from a completed swarm