Flower MCP

Overview Schema Source Related Servers Discussions Admin

fow_external_work

Destructive

Route revision-bound packets to external coding agents, export executable Markdown plans, derive dependency-ordered TODOs, and reconcile declared outcomes against issued authority.

Instructions

Hand a complete revision-bound packet to an external coding agent, export its executable Markdown plan, derive dependency-ordered TODOs and reconcile declared outcomes. Optional provider execution is a separate explicitly selected route. Route with operation: set_mode, select_executor, negotiate_provider, upgrade_validation, criteria, reconcile_criteria, add_criterion, edit_criterion, remove_criterion, validation, provider_validation, export_plan, todo, report_outcome, reconcile_outcome. Prerequisites: Use an existing packet/plan and issued authority. Select exactly one executor before execution. Effects: Reads derive criteria/readiness/TODO without executing code. Mutations record selection, criterion authority, artifacts and host outcomes; negotiate_provider alone explicitly probes optional provider support. Reporting completion does not verify or accept it. Exact help (replace operation_name with the selected route): fow_capabilities({"view":"operation","operation_tool":"fow_external_work","operation_name":"set_mode"}).

Input Schema

TableJSON Schema

Name	Required	Description
`actor`	No	Identity of the human or agent recording this action; retain the original identity when replaying its request.
`payload`	No	Standalone execution/PVP operation input. Copy issued spec/plan revisions and authority fingerprint where required; TODO/outcome identities and declared verification obligations must match the issued plan, including offline work. Applies to: set_mode, select_executor, negotiate_provider, upgrade_validation, reconcile_criteria, add_criterion, edit_criterion, remove_criterion, validation, provider_validation, export_plan, todo, report_outcome, reconcile_outcome. Required by: set_mode, select_executor, negotiate_provider, upgrade_validation, reconcile_criteria, add_criterion, edit_criterion, remove_criterion, export_plan, report_outcome, reconcile_outcome. Exact required/optional dotted fields and guidance: fow_capabilities(view=operation, operation_tool=fow_external_work, operation_name=<selected route>).
`change_id`	Yes	Canonical owning change ID; when address resolution permits omission, the packet/campaign must resolve unambiguously.
`operation`	Yes	Operation selector for this multiplexer; use a canonical route listed below. Canonical values: set_mode, select_executor, negotiate_provider, upgrade_validation, criteria, reconcile_criteria, add_criterion, edit_criterion, remove_criterion, validation, provider_validation, export_plan, todo, report_outcome, reconcile_outcome.
`packet_id`	Yes	Canonical packet identity returned by authoring; it identifies the bounded implementation work.
`project_id`	Yes	Canonical lifecycle project ID. Select the existing project explicitly; development Markdown authorities are separate from runtime projects.
`request_id`	No	Caller-owned retry identity. For receipted mutations replay the same actor and contents; changed contents can conflict. Inspect exact operation help for its receipt policy.

Output Schema

TableJSON Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

Changed9 schema fields changedv0.5.1Oct 16, 2026

addedInput schema / properties / actor / description

Added value: +"Identity of the human or agent recording this action; retain the original identity when replaying its request."

addedInput schema / properties / change_id / description

Added value: +"Canonical owning change ID; when address resolution permits omission, the packet/campaign must resolve unambiguously."

addedInput schema / properties / operation / description

Added value: +"Operation selector for this multiplexer; use a canonical route listed below. Canonical values: set_mode, select_executor, negotiate_provider, upgrade_validation, criteria, reconcile_criteria, add_criterion, edit_criterion, remove_criterion, validation, provider_validation, export_plan, todo, report_outcome, reconcile_outcome."

addedInput schema / properties / operation / examples
```
Added value: +[
+  "set_mode"
+]
```

addedInput schema / properties / packet_id / description

Added value: +"Canonical packet identity returned by authoring; it identifies the bounded implementation work."

addedInput schema / properties / payload / description

Added value: +"Standalone execution/PVP operation input. Copy issued spec/plan revisions and authority fingerprint where required; TODO/outcome identities and declared verification obligations must match the issued plan, including offline work. Applies to: set_mode, select_executor, negotiate_provider, upgrade_validation, reconcile_criteria, add_criterion, edit_criterion, remove_criterion, validation, provider_validation, export_plan, todo, report_outcome, reconcile_outcome. Required by: set_mode, select_executor, negotiate_provider, upgrade_validation, reconcile_criteria, add_criterion, edit_criterion, remove_criterion, export_plan, report_outcome, reconcile_outcome. Exact required/optional dotted fields and guidance: fow_capabilities(view=operation, operation_tool=fow_external_work, operation_name=<selected route>)."

addedInput schema / properties / payload / examples

Added value: +[
+  {
+    "expected_plan_revision": 1,
+    "expected_spec_revision": 1,
+    "mode": "external_agent"
+  }
+]

addedInput schema / properties / project_id / description

Added value: +"Canonical lifecycle project ID. Select the existing project explicitly; development Markdown authorities are separate from runtime projects."

addedInput schema / properties / request_id / description

Added value: +"Caller-owned retry identity. For receipted mutations replay the same actor and contents; changed contents can conflict. Inspect exact operation help for its receipt policy."

First observedv0.1.0Oct 16, 2026

TDQS

A3.5/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare destructiveHint=true, openWorldHint=true and non-idempotent, so the description's job is to add context. It does: reads derive criteria/readiness/TODO without executing code; mutations record authority and outputs; negotiate_provider is the only route that probes optional provider support; reporting completion does not verify or accept it. That last point is important behavioral context not covered by annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

It is front-loaded and organized with Route/Prerequisites/Effects paragraphs, which gives structure. However, the opening sentence is dense, the operation list duplicates the enum in the schema, and the help pointer is lengthy; the text is adequate but not tight.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists, so return values need not be explained. For a 7-parameter, 15-operation multiplexer the description covers routing, prerequisites and the key behavioral distinction between read and mutation routes, and points to fow_capabilities for per-operation field details. It is complete enough to call correctly, though per-operation semantics remain behind the capabilities tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, and the schema itself already describes actor, payload, change_id, packet_id, project_id, request_id and the operation enum. The description names the operations but adds little about individual parameter semantics beyond pointing to fow_capabilities for exact dotted fields. Baseline 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose3/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a broad summary of handing a packet to an external coding agent, exporting plans, deriving TODOs and reconciling outcomes, then lists 15 operations. That is a consistent descriptor for a multiplexer, so the purpose is not misleading, but it does not establish what fow_external_work owns relative to siblings like fow_packet_advance or fow_run. Its identity is only distinguishable by the operation list.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It states prerequisites (existing packet/plan and issued authority; exactly one executor selected before execution) and separates read operations from mutations. It does not tell the agent when to choose a sibling tool instead, but within this tool the read/mutation split and executor rule are clear routing guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.