nanobot/docs/my-tool.md
chengyongru 0bc42e2ab2 refactor: restrict fallback_models to preset-only and clean up provider factory
- Restrict fallback_models to only reference preset names in model_presets.
- Add schema validation to reject unknown preset names in fallback_models.
- Remove build_provider_for_model() since bare model fallback is no longer supported.
- Simplify make_provider_factory() to only look up presets by name.
- Update onboard UI to remove "Add custom model" option from fallback chain.
- Update tests to use preset names instead of bare model strings in fallback chains.
- Fix test imports referencing deleted _make_provider function.
2026-05-08 20:24:24 +08:00

7.8 KiB
Raw Blame History

My Tool

Let the agent sense and adjust its own runtime state — like asking a coworker "are you busy? can you switch to a bigger monitor?"

Why You Need It

Normal tools let the agent operate on the outside world (read/write files, search code). But the agent knows nothing about itself — it doesn't know which model it's running on, how many iterations are left, or how many tokens it has consumed.

My tool fills this gap. With it, the agent can:

  • Know who it is: What model am I using? Where is my workspace? How many iterations remain?
  • Adapt on the fly: Complex task? Expand the context window. Simple chat? Switch to a faster model.
  • Remember across turns: Store notes in your scratchpad that persist into the next conversation turn.

Note

This tool uses snake_case keys (model_preset, context_window_tokens). The matching config fields in config.json are camelCase (modelPreset, contextWindowTokens). See configuration.md for how to define presets in your config.

Configuration

Enabled by default (read-only mode). The agent can check its state but not set it.

tools:
  my:
    enable: true       # default: true
    allow_set: false   # default: false (read-only)

To allow the agent to set its configuration (e.g. switch models, adjust parameters), set tools.my.allow_set: true.

Legacy tools.myEnabled / tools.mySet keys are auto-migrated on load, and rewritten in-place the next time nanobot onboard refreshes the config.

All modifications are held in memory only — restart restores defaults.


check — Check "my" current state

Without parameters, returns a key config overview:

my(action="check")
# → max_iterations: 40
#   model_preset: 'fast'
#   workspace: PosixPath('/tmp/workspace')
#   provider_retry_mode: 'standard'
#   max_tool_result_chars: 16000
#   _current_iteration: 3
#   _last_usage: {'prompt_tokens': 45000, 'completion_tokens': 8000}
#   Note: prompt_tokens is cumulative across all turns, not current context window occupancy.

With a key parameter, drill into a specific config:

my(action="check", key="_last_usage.prompt_tokens")
# → How many prompt tokens I've used so far

my(action="check", key="model_preset")
# → Current active preset name (e.g. 'fast')

my(action="check", key="model_presets")
# → Lists all preset names and their models, e.g.:
#   fast → gpt-4.1-mini (openai)
#   deep → claude-opus-4-7 (anthropic)

my(action="check", key="web_config.enable")
# → Whether web search is enabled

What you can do with it

Scenario How
"What model are you using?" check("model_preset")
"How many more tool calls can you make?" check("max_iterations") minus check("_current_iteration")
"How many tokens has this conversation used?" check("_last_usage") — cumulative across all turns
"Where is your working directory?" check("workspace")
"Show me your full config" check()
"Are there any subagents running?" check("subagents") — shows phase, iteration, elapsed time, tool events

set — Runtime tuning

Changes take effect immediately, no restart required.

my(action="set", key="max_iterations", value=80)
# → Bump iteration limit from 40 to 80

my(action="set", key="model_preset", value="fast")
# → Switch to the 'fast' preset (model, provider, temperature, etc. all at once)
#
# If the preset name does not exist:
# → Error: model_preset 'unknown' not found. Available: fast, deep

my(action="set", key="context_window_tokens", value=131072)
# → Expand context window for long documents

You can also store custom state in your scratchpad:

my(action="set", key="current_project", value="nanobot")
my(action="set", key="user_style_preference", value="concise")
my(action="set", key="task_complexity", value="high")
# → These values persist into the next conversation turn

Protected parameters

These parameters have validation — invalid values are rejected:

Parameter Type Range / Constraint Purpose
max_iterations int 1100 Max tool calls per conversation turn
model_preset str must exist in model_presets Switch to a named preset bundle

Other parameters (e.g. model, context_window_tokens, workspace, provider_retry_mode, max_tool_result_chars) can be set freely, as long as the value is JSON-safe.

Note

Setting model or context_window_tokens directly automatically clears the active model_preset, because the live state no longer matches the preset bundle. Use model_preset for atomic switches instead.


Practical Scenarios

"This task is complex, I need more room"

Agent: This codebase is large, let me expand my context window to handle it.
→ my(action="set", key="context_window_tokens", value=131072)

"Simple question, don't waste compute"

Agent: This is a straightforward question, let me switch to the fast preset.
→ my(action="set", key="model_preset", value="fast")

"Remember user preferences across turns"

Turn 1: my(action="set", key="user_prefers_concise", value=True)
Turn 2: my(action="check", key="user_prefers_concise")
# → True (still remembers the user likes concise replies)

"Self-diagnosis"

User: "Why aren't you searching the web?"
Agent: Let me check my web config.
→ my(action="check", key="web_config.enable")
# → False
Agent: Web search is disabled — please set web.enable: true in your config.

"Token budget management"

Agent: Let me check how much budget I have left.
→ my(action="check", key="_last_usage")
# → {"prompt_tokens": 45000, "completion_tokens": 8000}
Agent: I've used ~53k tokens total so far. I'll keep my remaining replies concise.

"Subagent monitoring"

Agent: Let me check on the background tasks.
→ my(action="check", key="subagents")
# → 2 subagent(s):
#   [task-1] 'Code review'
#     phase: running, iteration: 5, elapsed: 12.3s
#     tools: read(✓), grep(✓)
#     usage: {'prompt_tokens': 8000, 'completion_tokens': 1200}
#   [task-2] 'Write tests'
#     phase: pending, iteration: 0, elapsed: 0.2s
#     tools: none
Agent: The code review is progressing well. The test task hasn't started yet.

Safety Mechanisms

Core design principle: All modifications live in memory only. Restart restores defaults. The agent cannot cause persistent damage.

Off-limits (BLOCKED)

Cannot be checked or modified — fully hidden:

Category Attributes Reason
Core infrastructure bus, provider, _running Changes would crash the system
Tool registry tools Must not remove its own tools
Subsystems runner, sessions, consolidator, etc. Affects other users/sessions
Sensitive data _mcp_servers, _pending_queues, etc. Contains credentials and message routing
Security boundaries restrict_to_workspace, channels_config Bypassing would violate isolation
Python internals __class__, __dict__, etc. Prevents sandbox escape

Read-only (check only)

Can be checked but not set:

Category Attributes Reason
Subagent manager subagents Observable, but replacing breaks the system
Execution config exec_config Can check sandbox/enable status, cannot change it
Web config web_config Can check enable status, cannot change it
Iteration counter _current_iteration Updated by runner only

Sensitive field protection

Sub-fields matching sensitive names (api_key, password, secret, token, etc.) are blocked from both check and set, regardless of parent path. This prevents credential leaks via dot-path traversal (e.g. web_config.search.api_key).