list-learned-actions

List learned actions matching: $ARGUMENTS (optional keyword filter; omit to list all)

Why this command exists

Past sessions accumulate two kinds of reusable knowledge that future sessions should look at BEFORE re-deriving anything from scratch:

1. Feedback memories at ~/.claude/projects/<encoded-cwd>/memory/feedback_*.md
2. Executable artifacts: .rn-agent/actions/*.yaml and *.yml, .rn-agent/skeleton.yaml

This command surfaces both lists in one place — and the underlying packages/rn-dev-agent-core/src/learned-actions.ts (compiled to packages/rn-dev-agent-core/dist/learned-actions.js) is the same script invoked programmatically by /rn-dev-agent:test-feature Step 0 and by the rn-tester / rn-debugger agents' artifact-scan steps. Keeping the discovery logic in one script means every consumer sees the same inventory.

Run

node "${CLAUDE_PLUGIN_ROOT}/rn-dev-agent-core/dist/learned-actions.js" \
  --workspace-root "$PWD" \
  --memory-cwd "$PWD" \
  ${ARGUMENTS:+--filter "$ARGUMENTS"}

The script auto-discovers:

• The user's per-project auto-memory directory (Section A)
• .rn-agent/actions/*.yaml and *.yml in the cwd, in <cwd>/test-app/, and in any <sibling>/test-app/ adjacent to the cwd (Section B)
• .rn-agent/skeleton.yaml in the same locations (Section C)
• Plugin commands (Section D — only populated when running inside the plugin repo)

It exits 0 when results found, 3 when nothing matches, 2 on bad flags.

Programmatic invocation

Other commands and agents can call the script directly to get JSON for decision-making. Always use --json from a programmatic caller — the human table format is not a stable contract.

RESULT=$(node "${CLAUDE_PLUGIN_ROOT}/rn-dev-agent-core/dist/learned-actions.js" \
  --json --section b --filter "task creation" --workspace-root "$PWD" --memory-cwd "$PWD")
echo "$RESULT" | jq '.sections.flows.items[0].path'

JSON shape (abridged):

{
  "cwd": "/Users/.../my-rn-app",
  "filter": "cart",
  "sections": {
    "memories": { "count": 0, "dir": "...", "items": [] },
    "flows": {
      "count": 1,
      "roots": ["..."],
      "items": [
        {
          "flow": "cart-add-item",
          "path": "/.../cart-add-item.yaml",
          "appId": "com.example.app",
          "purpose": "Add an item to the cart and verify the badge increments",
          "params": ["ITEM_ID", "QTY"],
          "replay": "display-only legacy command; execute through cdp_run_action"
        }
      ]
    },
    "skeletons": { "count": 1, "items": [...] },
    "commands": { "count": 16, "items": [...] }
  },
  "total": 18
}