fileforge_task_action

Previous value: -"The action to perform. Available actions: start: Start an asynchronous task made of structured FileForge tool invocations. add: Add a saved task made of structured FileForge tool invocations. edit: Edit a saved task record. status: Get the current status of a running task. events: Get progress events for a task. pause: Pause a running task. resume: Resume a paused task. cancel: Cancel a running task. list: List all tasks (running, paused, completed). delete: Delete a task and its event history. export: Export task results and artifacts. monitor_page: Monitor a page for changes over time. collect_table: Collect rows from a paginated table into a local dataset. collect_search_results: Run a search on a website and collect all results into a dataset. create: Create a routine (prompt or steps input). update: Update a routine (snapshots a new version). get: Fetch a routine (optionally a specific version). duplicate: Duplicate an existing routine. enable: Enable a routine. disable: Disable a routine. validate: Validate a routine or an arbitrary step array against the tool catalog. dry_run: Produce a dry-run report for a routine without executing anything. run: Enqueue a routine for execution. stop: Stop / cancel a running or queued run. get_runs: List runs for a routine. get_run_details: Get full details of a run (steps + events + artifacts). export_run: Export a run as JSON or CSV. get_tool_registry: Return the FileForge tool capability catalog. plan_from_prompt: Return planning context (registry slice + schema + planning prompt) for an LLM to use when converting a prompt into an executable step array. commit_plan: Validate an LLM-produced step array and save it as a routine (new or updated)."New value: +"What to do. Required parameters are in brackets. start: Start an asynchronous task made of structured FileForge tool invocations. add (actions): Add a saved task made of structured FileForge tool invocations. edit (taskId): Edit a saved task record. status (taskId): Get the current status of a running task. events (taskId): Get progress events for a task. pause (taskId): Pause a running task. resume (taskId): Resume a paused task. cancel (taskId): Cancel a running task. list: List all tasks (running, paused, completed). delete (taskId): Delete a task and its event history. export (taskId): Export task results and artifacts. monitor_page (url): Monitor a page for changes over time. collect_table (url): Collect rows from a paginated table into a local dataset. collect_search_results (url, query): Run a search on a website and collect all results into a dataset. create (name): Create a routine (prompt or steps input). update (routineId, patch): Update a routine (snapshots a new version). get (routineId): Fetch a routine (optionally a specific version). duplicate (routineId): Duplicate an existing routine. enable (routineId): Enable a routine. disable (routineId): Disable a routine. validate: Validate a routine or an arbitrary step array against the tool catalog. dry_run (routineId): Produce a dry-run report for a routine without executing anything. run (routineId): Enqueue a routine for execution. stop (runId): Stop / cancel a running or queued run. get_runs (routineId): List runs for a routine. get_run_details (runId): Get full details of a run (steps + events + artifacts). export_run (runId): Export a run as JSON or CSV. get_tool_registry: Return the FileForge tool capability catalog. plan_from_prompt (prompt): Return planning context (registry slice + schema + planning prompt) for an LLM to use when converting a prompt into an executable step array. commit_plan (plannedSteps): Validate an LLM-produced step array and save it as a routine (new or updated)."

Added value: +"Page sections to scrape while running."

Added value: +"What to do when a run is already active."

Added value: +"Who created it."

Added value: +"Cron schedule for scheduled routines."

Added value: +"Default input values."

Added value: +"Whether the routine can run."

Removed value: -{
-  "items": {
-    "type": "string"
-  },
-  "type": "array"
-}

Added value: +"Saved with the routine. Runs do not act on it yet."

Added value: +"Output format."

Added value: +"Extra guidance for plan_from_prompt."

Added value: +"Key that prevents starting the same run twice."

Added value: +"Include run events."

Added value: +"Include logs."

Added value: +"Include raw output."

Added value: +"Include each step's result."

Added value: +"Include a summary."

Added value: +"Whether the task starts from a prompt or from steps."

Added value: +"Minutes between runs."

Added value: +"Maximum items to return."

Added value: +"Maximum run time in minutes."

Added value: +"Maximum pages to visit."

Added value: +"Free-form metadata to store with the routine."

Added value: +"Kind of task: interactive, collector or monitor."

Added value: +"Routine name."

Added value: +"Saved with the routine. Runs do not act on it yet."

Added value: +"Items to skip, for paging."

Removed value: -{
-  "items": {
-    "type": "object"
-  },
-  "type": "array"
-}

Added value: +"ID of the parent run."

Added value: +"Routine fields to change, using the same names as create."

Added value: +"What the routine may do, such as { allowBrowser: true }. Also allowNetwork, allowScreenshots, allowCookies, allowClipboard, allowDownloads, allowFilesystemRead, allowFilesystemWrite, allowAuthenticatedSession, allowExternalWebhooks."

Added value: +"Steps from plan_from_prompt to commit."

Added value: +"Milliseconds between polls."

Added value: +"Queue priority. Higher runs first."

Added value: +"Plain-language description of the routine."

Added value: +"Input values for this run."

Added value: +"JSON schema for the routine's inputs."

Added value: +"Search text to collect results for."

Added value: +"Why the change was made."

Added value: +"Result data to store on the task."

Added value: +"Saved with the routine. Runs do not act on it yet."

Added value: +"ID of the run this one retries."

Added value: +"{ count, backoffMs } for failed steps. A step's own retry takes priority."

Added value: +"Routine ID."

Added value: +"Run ID."

Added value: +"Simulate unavailable tools during validation."

Added value: +"add and edit: pending, running, paused, completed, failed or cancelled. list: any of those, or \"all\". get_runs: a run status."

Added value: +"Ordered steps: { id, tool, params, dependsOn, condition, onFailure, retry, timeout, outputKey, requiresApproval }. params includes the tool's action."

Added value: +"Saved with the routine. Runs do not act on it yet."

Added value: +"Save the result in ~/.fileforge and return a ref."

Added value: +"{ stepMs }: time limit per step. Defaults to 60000."

Added value: +"Time limit in milliseconds."

Added value: +"What started this run."

Added value: +"What starts the routine."

Added value: +"Who made the change."

Added value: +"Routine version."

Previous value: -nullNew value: +{
+  "description": "query_* actions return their modules as extra top-level fields.",
+  "properties": {
+    "data": {
+      "description": "The result. Its shape depends on the action."
+    },
+    "error": {
+      "description": "Failure code and message."
+    },
+    "meta": {
+      "description": "Tool, action and context."
+    },
+    "ok": {
+      "description": "False if the action failed or a check did not pass.",
+      "type": "boolean"
+    },
+    "ref": {
+      "description": "Stored result ID, for fileforge_result_action."
+    },
+    "warnings": {
+      "description": "Notes about a call that still ran."
+    }
+  },
+  "required": [
+    "ok"
+  ],
+  "type": "object"
+}