Skip to content

borghei/Claude-Skills / engineering/self-improving-agent

self-improving-agent

Verified

Curates agent memory and turns repeated lessons into reviewed rules: memory health checks, pattern extraction, feedback and regression analysis. Use when managing auto-memory or building agents that learn from their own sessions.

assets/sample_metrics.json

View on GitHub
{
  "baseline": {
    "period": "2026-02-01 to 2026-02-15",
    "sessions": [
      {
        "session_id": "b0",
        "task_type": "code-review",
        "outcome": "SUCCESS",
        "corrections": 0,
        "turns": 4,
        "tool_calls": 12,
        "tool_errors": 0,
        "context_items_retrieved": 5,
        "context_items_used": 4
      },
      {
        "session_id": "b1",
        "task_type": "code-review",
        "outcome": "SUCCESS",
        "corrections": 0,
        "turns": 4,
        "tool_calls": 12,
        "tool_errors": 0,
        "context_items_retrieved": 5,
        "context_items_used": 3
      },
      {
        "session_id": "b2",
        "task_type": "code-review",
        "outcome": "SUCCESS",
        "corrections": 0,
        "turns": 4,
        "tool_calls": 12,
        "tool_errors": 0,
        "context_items_retrieved": 5,
        "context_items_used": 4
      },
      {
        "session_id": "b3",
        "task_type": "code-review",
        "outcome": "SUCCESS",
        "corrections": 0,
        "turns": 4,
        "tool_calls": 12,
        "tool_errors": 0,
        "context_items_retrieved": 5,
        "context_items_used": 3
      },
      {
        "session_id": "b4",
        "task_type": "code-review",
        "outcome": "SUCCESS",
        "corrections": 0,
        "turns": 4,
        "tool_calls": 12,
        "tool_errors": 0,
        "context_items_retrieved": 5,
        "context_items_used": 4
      },
      {
        "session_id": "b5",
        "task_type": "code-review",
        "outcome": "SUCCESS",
        "corrections": 0,
        "turns": 4,
        "tool_calls": 12,
        "tool_errors": 0,
        "context_items_retrieved": 5,
        "context_items_used": 3
      }
    ]
  },
  "current": {
    "period": "2026-03-01 to 2026-03-15",
    "sessions": [
      {
        "session_id": "c0",
        "task_type": "code-review",
        "outcome": "FAILURE",
        "corrections": 1,
        "turns": 7,
        "tool_calls": 12,
        "tool_errors": 0,
        "context_items_retrieved": 5,
        "context_items_used": 4
      },
      {
        "session_id": "c1",
        "task_type": "code-review",
        "outcome": "SUCCESS",
        "corrections": 2,
        "turns": 7,
        "tool_calls": 12,
        "tool_errors": 1,
        "context_items_retrieved": 5,
        "context_items_used": 3
      },
      {
        "session_id": "c2",
        "task_type": "code-review",
        "outcome": "SUCCESS",
        "corrections": 1,
        "turns": 7,
        "tool_calls": 12,
        "tool_errors": 2,
        "context_items_retrieved": 5,
        "context_items_used": 4
      },
      {
        "session_id": "c3",
        "task_type": "code-review",
        "outcome": "FAILURE",
        "corrections": 2,
        "turns": 7,
        "tool_calls": 12,
        "tool_errors": 0,
        "context_items_retrieved": 5,
        "context_items_used": 3
      },
      {
        "session_id": "c4",
        "task_type": "code-review",
        "outcome": "SUCCESS",
        "corrections": 1,
        "turns": 7,
        "tool_calls": 12,
        "tool_errors": 1,
        "context_items_retrieved": 5,
        "context_items_used": 4
      },
      {
        "session_id": "c5",
        "task_type": "code-review",
        "outcome": "SUCCESS",
        "corrections": 2,
        "turns": 7,
        "tool_calls": 12,
        "tool_errors": 2,
        "context_items_retrieved": 5,
        "context_items_used": 3
      }
    ]
  },
  "changes": [
    {
      "date": "2026-02-20",
      "type": "rule_added",
      "description": "Always run the full test suite before review"
    }
  ]
}
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: 471kUpdated 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