moltbot/llm-task.md at 82c4fd8f56751faa03470e32f4763b7245c69b71

mirror of https://github.com/moltbot/moltbot.git synced 2026-05-07 07:58:36 +00:00

Files

Peter Steinberger e5ec14a06a fix(plugins): discover alsoAllow plugin tools

Summary:
- Discover optional plugin tools named in tools.alsoAllow without treating additive alsoAllow as a restrictive plugin-tool allowlist.
- Preserve explicit alsoAllow wildcards and keep default non-optional plugin tools visible.
- Document llm-task and lobster enablement and add changelog coverage.

Verification:
- pnpm test src/agents/tool-policy.test.ts src/gateway/tools-invoke-http.test.ts src/agents/pi-tools.create-openclaw-coding-tools.test.ts src/plugins/tools.optional.test.ts
- pnpm exec oxfmt --check --threads=1 src/agents/sandbox-tool-policy.ts src/agents/tool-policy.ts src/agents/tool-policy.test.ts src/agents/pi-tools.create-openclaw-coding-tools.test.ts src/gateway/tools-invoke-http.test.ts src/plugins/tools.ts src/plugins/tools.optional.test.ts
- git diff --check
- Blacksmith Testbox tbx_01kqr05924hz9kw50myxrqmsf9: pnpm check:changed

Fixes #76616

2026-05-03 23:46:14 +01:00

2.7 KiB

Raw Blame History

summary, read_when, title

summary

read_when

title

JSON-only LLM tasks for workflows (optional plugin tool)

You want a JSON-only LLM step inside workflows

You need schema-validated LLM output for automation

LLM task

llm-task is an optional plugin tool that runs a JSON-only LLM task and returns structured output (optionally validated against JSON Schema).

This is ideal for workflow engines like Lobster: you can add a single LLM step without writing custom OpenClaw code for each workflow.

Enable the plugin

Enable the plugin:

{
  "plugins": {
    "entries": {
      "llm-task": { "enabled": true }
    }
  }
}

Allow the optional tool:

{
  "tools": {
    "alsoAllow": ["llm-task"]
  }
}

Use tools.allow only when you want restrictive allowlist mode.

Config (optional)

{
  "plugins": {
    "entries": {
      "llm-task": {
        "enabled": true,
        "config": {
          "defaultProvider": "openai-codex",
          "defaultModel": "gpt-5.5",
          "defaultAuthProfileId": "main",
          "allowedModels": ["openai/gpt-5.4"],
          "maxTokens": 800,
          "timeoutMs": 30000
        }
      }
    }
  }
}

allowedModels is an allowlist of provider/model strings. If set, any request outside the list is rejected.

Tool parameters

prompt (string, required)
input (any, optional)
schema (object, optional JSON Schema)
provider (string, optional)
model (string, optional)
thinking (string, optional)
authProfileId (string, optional)
temperature (number, optional)
maxTokens (number, optional)
timeoutMs (number, optional)

thinking accepts the standard OpenClaw reasoning presets, such as low or medium.

Output

Returns details.json containing the parsed JSON (and validates against schema when provided).

Example: Lobster workflow step

openclaw.invoke --tool llm-task --action json --args-json '{
  "prompt": "Given the input email, return intent and draft.",
  "thinking": "low",
  "input": {
    "subject": "Hello",
    "body": "Can you help?"
  },
  "schema": {
    "type": "object",
    "properties": {
      "intent": { "type": "string" },
      "draft": { "type": "string" }
    },
    "required": ["intent", "draft"],
    "additionalProperties": false
  }
}'

Safety notes

The tool is JSON-only and instructs the model to output only JSON (no code fences, no commentary).
No tools are exposed to the model for this run.
Treat output as untrusted unless you validate with schema.
Put approvals before any side-effecting step (send, post, exec).

2.7 KiB Raw Blame History