Skip to content

LeoYeAI/agent-master-skills / skills/agent-harness

agent-harness

Verified

Use when an agent should drive a goal in one of this repo's domains to a verified close, before starting the work. Compiles the goal into a task plan with goal_compiler.py and runs it through loop_controller.py, which verifies each task, caps retries and escalates.

assets/harnesses/research-ops.json

View on GitHub
{
  "schema": "agent-harness/manifest.v1",
  "domain": "research-ops",
  "skill_count": 6,
  "loop_defaults": {
    "max_attempts_per_task": 3,
    "max_loop_iterations": 12,
    "escalate_on": [
      "attempts_exhausted",
      "no_verification_available",
      "destructive_or_irreversible_action",
      "goal_drift_detected"
    ]
  },
  "skills": [
    {
      "name": "clinical-research",
      "path": "research-ops/skills/clinical-research",
      "description": "Use when designing a prospective clinical study: choosing primary and secondary endpoints, estimating sample size and power, or scoring feasibility before submission. Runs endpoint_selector.py, sample_size_estimator.py and phase_gate_scorer.py; outputs are estimates for a biostatistician to confirm.",
      "tools": [
        {
          "script": "research-ops/skills/clinical-research/scripts/ar_evaluator.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/clinical-research/scripts/ar_evaluator.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/clinical-research/scripts/ar_evaluator.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/clinical-research/scripts/config_loader.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/clinical-research/scripts/config_loader.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/clinical-research/scripts/config_loader.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/clinical-research/scripts/endpoint_selector.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/clinical-research/scripts/endpoint_selector.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/clinical-research/scripts/endpoint_selector.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/clinical-research/scripts/onboard.py",
          "wired": true,
          "supports_sample": false,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/clinical-research/scripts/onboard.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            }
          ]
        },
        {
          "script": "research-ops/skills/clinical-research/scripts/phase_gate_scorer.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/clinical-research/scripts/phase_gate_scorer.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/clinical-research/scripts/phase_gate_scorer.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/clinical-research/scripts/sample_size_estimator.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/clinical-research/scripts/sample_size_estimator.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/clinical-research/scripts/sample_size_estimator.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        }
      ],
      "agentic_signals": {
        "goal_intake": true,
        "refusal_gate": false,
        "verification": true,
        "loop_discipline": true,
        "close_out": false
      },
      "references": [
        "endpoint_and_power.md",
        "study_design_canon.md",
        "trial_operations.md"
      ]
    },
    {
      "name": "interview-mapper",
      "path": "research-ops/skills/interview-mapper",
      "description": "Use when coding or making sense of one interview transcript (customer, JTBD, expert, usability, exit). Extracts verbatim-grounded nuggets with extract_nuggets.py, verifies quotes with verify_quotes.py and checks conclusions hold across reruns with consensus.py.",
      "tools": [
        {
          "script": "research-ops/skills/interview-mapper/scripts/batch_prepare.py",
          "wired": false,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/batch_prepare.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/batch_prepare.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/build_provenance.py",
          "wired": false,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/build_provenance.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/build_provenance.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/calibrate_threshold.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/calibrate_threshold.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/calibrate_threshold.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/check_support.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/check_support.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/check_support.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/compare_to_gold.py",
          "wired": false,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/compare_to_gold.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/compare_to_gold.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/consensus.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/consensus.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/consensus.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/coverage_gaps.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/coverage_gaps.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/coverage_gaps.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/extract_claims.py",
          "wired": false,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/extract_claims.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/extract_claims.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/extract_nuggets.py",
          "wired": false,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/extract_nuggets.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/extract_nuggets.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/make_adjudication.py",
          "wired": false,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/make_adjudication.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/make_adjudication.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/number_lines.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/number_lines.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/number_lines.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/render_board.py",
          "wired": false,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/render_board.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/render_board.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/route.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/route.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/route.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/score_insights.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/score_insights.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/score_insights.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/interview-mapper/scripts/verify_quotes.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/verify_quotes.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/interview-mapper/scripts/verify_quotes.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        }
      ],
      "agentic_signals": {
        "goal_intake": true,
        "refusal_gate": false,
        "verification": true,
        "loop_discipline": false,
        "close_out": false
      },
      "references": [
        "ethics.md",
        "intake.md",
        "pipeline.md",
        "reliability.md",
        "rubric.md",
        "synthesis.md",
        "validation.md"
      ]
    },
    {
      "name": "market-research",
      "path": "research-ops/skills/market-research",
      "description": "Use for market sizing (TAM, SAM, SOM both top-down and bottom-up), survey sample sizes or segment scoring. Runs market_sizer.py, sample_size_planner.py and segmentation_scorer.py and never returns one unsourced number.",
      "tools": [
        {
          "script": "research-ops/skills/market-research/scripts/ar_evaluator.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/market-research/scripts/ar_evaluator.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/market-research/scripts/ar_evaluator.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/market-research/scripts/config_loader.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/market-research/scripts/config_loader.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/market-research/scripts/config_loader.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/market-research/scripts/market_sizer.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/market-research/scripts/market_sizer.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/market-research/scripts/market_sizer.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/market-research/scripts/onboard.py",
          "wired": true,
          "supports_sample": false,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/market-research/scripts/onboard.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            }
          ]
        },
        {
          "script": "research-ops/skills/market-research/scripts/sample_size_planner.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/market-research/scripts/sample_size_planner.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/market-research/scripts/sample_size_planner.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/market-research/scripts/segmentation_scorer.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/market-research/scripts/segmentation_scorer.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/market-research/scripts/segmentation_scorer.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        }
      ],
      "agentic_signals": {
        "goal_intake": true,
        "refusal_gate": false,
        "verification": true,
        "loop_discipline": false,
        "close_out": false
      },
      "references": [
        "market_sizing_canon.md",
        "segmentation_and_ci.md",
        "survey_methodology.md"
      ]
    },
    {
      "name": "product-research",
      "path": "research-ops/skills/product-research",
      "description": "Use when planning or synthesizing user research: choosing the method, deciding how many participants, or turning coded notes into insights. Runs study_designer.py, saturation_planner.py and insight_synthesizer.py, which refuses single-source insights.",
      "tools": [
        {
          "script": "research-ops/skills/product-research/scripts/ar_evaluator.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/product-research/scripts/ar_evaluator.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/product-research/scripts/ar_evaluator.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/product-research/scripts/config_loader.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/product-research/scripts/config_loader.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/product-research/scripts/config_loader.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/product-research/scripts/insight_synthesizer.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/product-research/scripts/insight_synthesizer.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/product-research/scripts/insight_synthesizer.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/product-research/scripts/onboard.py",
          "wired": true,
          "supports_sample": false,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/product-research/scripts/onboard.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            }
          ]
        },
        {
          "script": "research-ops/skills/product-research/scripts/saturation_planner.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/product-research/scripts/saturation_planner.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/product-research/scripts/saturation_planner.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/product-research/scripts/study_designer.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/product-research/scripts/study_designer.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/product-research/scripts/study_designer.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        }
      ],
      "agentic_signals": {
        "goal_intake": true,
        "refusal_gate": false,
        "verification": true,
        "loop_discipline": false,
        "close_out": false
      },
      "references": [
        "repository_and_synthesis.md",
        "research_methods_canon.md",
        "sampling_and_saturation.md"
      ]
    },
    {
      "name": "research-finance",
      "path": "research-ops/skills/research-finance",
      "description": "Use for R&D program money: multi-period budgets with indirect cost split, burn and runway against milestones, or capitalize vs expense routing. Runs program_budget_planner.py, burn_runway_tracker.py and capex_vs_opex_router.py.",
      "tools": [
        {
          "script": "research-ops/skills/research-finance/scripts/ar_evaluator.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/research-finance/scripts/ar_evaluator.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/research-finance/scripts/ar_evaluator.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/research-finance/scripts/burn_runway_tracker.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/research-finance/scripts/burn_runway_tracker.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/research-finance/scripts/burn_runway_tracker.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/research-finance/scripts/capex_vs_opex_router.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/research-finance/scripts/capex_vs_opex_router.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/research-finance/scripts/capex_vs_opex_router.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/research-finance/scripts/config_loader.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/research-finance/scripts/config_loader.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/research-finance/scripts/config_loader.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        },
        {
          "script": "research-ops/skills/research-finance/scripts/onboard.py",
          "wired": true,
          "supports_sample": false,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/research-finance/scripts/onboard.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            }
          ]
        },
        {
          "script": "research-ops/skills/research-finance/scripts/program_budget_planner.py",
          "wired": true,
          "supports_sample": true,
          "verification": [
            {
              "cmd": "python3 research-ops/skills/research-finance/scripts/program_budget_planner.py --help",
              "expect_exit": 0,
              "kind": "smoke"
            },
            {
              "cmd": "python3 research-ops/skills/research-finance/scripts/program_budget_planner.py --sample",
              "expect_exit": 0,
              "kind": "sample"
            }
          ]
        }
      ],
      "agentic_signals": {
        "goal_intake": true,
        "refusal_gate": false,
        "verification": true,
        "loop_discipline": true,
        "close_out": false
      },
      "references": [
        "burn_and_portfolio.md",
        "indirect_rate_modeling.md",
        "rd_program_finance_canon.md"
      ]
    },
    {
      "name": "research-ops-skills",
      "path": "research-ops/skills/research-ops-skills",
      "description": "Use when an enterprise research request needs routing (clinical study design, R&D finance, market sizing, user research): 'what sample size', 'R&D budget'. Lists the research-ops skills and routes to one.",
      "tools": [],
      "agentic_signals": {
        "goal_intake": true,
        "refusal_gate": false,
        "verification": false,
        "loop_discipline": true,
        "close_out": true
      },
      "references": []
    }
  ]
}
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall