Requirements:

• I want to have research agent (web) that is available for all project, sothat I can do structured web search on topics.
• The research agent should store it's research memory as text (.md) files in local project folder structure for to create growing context for the project. ## Custom Modes and Agents

Kilo discovers these files via glob patterns in multiple config directories:

Scope Mode file path pattern Agent file path pattern

Project-local .kilo/modes/.md .kilo/agents/.md

Global (XDG) ~/.config/kilo/modes/.md ~/.config/agents/modes/.md

Legacy global ~/.kilo/modes/.md or ~/.kilocode/modes/.md

• https://developers.openai.com/cookbook/examples/agents_sdk/building_reliable_agents_memory_compaction
• https://github.com/Kilo-Org/kilocode/discussions/2022
• https://github.com/Kilo-Org/kilocode/discussions/377

Folder and Manifest Best Practices

Put source documents, manifests, helper files, and output directories in the Manifest instead of pasting large content into the prompt.

Put longer task instructions in workspace files such as README.md, task.md, or AGENTS.md; keep agent instructions focused on behavior and boundaries.

Use stable document IDs and a machine-readable manifest file so generated memos can cite sources and reviewers can inspect the path back to evidence.

Let Memory() manage its own memory artifacts. By default, sandbox memory uses memories/ and sessions/ under the workspace.

Keep generated artifacts under outputs/ so the application can inspect, copy, validate, or archive them after the run.

Keep mount scopes narrow. If you mount a data room, mount only what the agent should read or write.

Treat secrets as runtime configuration injected by your application or sandbox provider, not as prompt text or committed manifest content.

Prefer a small synthetic File(…) or Dir(…) entry for a tutorial, then switch to LocalDir, GitRepo, or storage mounts for production-sized datasets .

WORKSPACE_TREE = """
/workspace/
  README.md
  manifest.csv
  docs/
    batch_1/
    batch_2/
    batch_3/
  outputs/
  memories/    # Generated by Memory()
  sessions/    # Generated by Memory()
""".strip()

print(WORKSPACE_TREE)

Prepare a Small Evidence Workspace