harness-forge plugin
Native Meta-Harness — run a harness-optimization loop natively: propose scaffolding variants around a fixed model, score each with a $0 deterministic test, keep the Pareto-best, loop.
How to install the harness-forge plugin
Run both commands inside Claude Code. The first adds the marketplace once; the second installs the plugin.
/plugin marketplace add 001TMF/harness-forge
/plugin install harness-forge@tmf-skillsWhat's inside harness-forge
1 part read from the plugin's own files.
Skills 1
Claude loads it when the task matches- meta-harness
Run a Meta-Harness-style optimization loop NATIVELY — automatically search over the scaffolding around a FIXED base model (memory, retrieval, context construction, prompt templates, summarization, tool-selection logic) by proposing candidate variants, scoring each on a cheap deterministic eval, and keeping a Pareto frontier of quality vs cost — using native Agent / Workflow / loop tools instead of a standalone Python harness. Use this whenever the user wants to optimize, evolve, tune, distill, or search over a harness, scaffold, prompt system, memory or retrieval policy, context-assembly code, or summarizer while keeping the model fixed; whenever they mention Meta-Harness, harness optimization, scaffold evolution, automatic prompt/memory optimization, an evolutionary or Pareto search over candidate implementations, or "make the harness/agent better without retraining"; and whenever the gain must come from the code AROUND the model rather than the model weights. Reproduces the Meta-Harness paper's method nativ
Using harness-forge in Claude Code
Skills need no command. Claude reads them when your request matches what they describe.
Update, turn off or remove
- Update the plugin (shell)
- claude plugin update harness-forge@tmf-skills
- Refresh the marketplace
- /plugin marketplace update tmf-skills
- Turn it off for now
- /plugin disable harness-forge@tmf-skills
- Remove it
- /plugin uninstall harness-forge@tmf-skills
Skills in harness-forge
- meta-harness001TMF/harness-forge
Run a Meta-Harness-style optimization loop NATIVELY — automatically search over the scaffolding around a FIXED base model (memory, retrieval, context construction, prompt templates, summarization, tool-selection logic) by proposing candidate variants, scoring each on a cheap deterministic eval, and keeping a Pareto frontier of quality vs cost — using native Agent / Workflow / loop tools instead of a standalone Python harness. Use this whenever the user wants to optimize, evolve, tune, distill, or search over a harness, scaffold, prompt system, memory or retrieval policy, context-assembly code, or summarizer while keeping the model fixed; whenever they mention Meta-Harness, harness optimization, scaffold evolution, automatic prompt/memory optimization, an evolutionary or Pareto search over candidate implementations, or "make the harness/agent better without retraining"; and whenever the gain must come from the code AROUND the model rather than the model weights. Reproduces the Meta-Harness paper's method nativ
10 files81stars
harness-forge plugin questions
How do I install the harness-forge plugin?
In Claude Code run /plugin marketplace add 001TMF/harness-forge, then /plugin install harness-forge@tmf-skills.
What does the harness-forge plugin include?
harness-forge bundles 1 skill.
How do I update or remove the harness-forge plugin?
Update it with claude plugin update harness-forge@tmf-skills in your shell, or from the Installed tab of /plugin. Remove it with /plugin uninstall harness-forge@tmf-skills.
What license is the harness-forge plugin under?
MIT, as stated in the plugin's manifest.