flow-workflows 0.71.0 → 0.72.0

package/adapters/codex/CORE.md

@@ -17,7 +17,7 @@ Every `$flow-*` command assumes these rules. They are stated once, here, so a co
 carries what is specific to its phase. Read this once per session; a command that says "load
 `flow-core`" means this file.
 
-**This file belongs to flow `0.71.0`.** Compare it once, at the start of the session, against
+**This file belongs to flow `0.72.0`.** Compare it once, at the start of the session, against
 `version` in `~/.claude/flow/.claude-plugin/plugin.json`. The two differing means the session
 is running a **mixture** — the commands from one copy of the plugin, these shared rules from another
 — which is exactly what happens when a branch or a release candidate is loaded over an installed
@@ -269,8 +269,17 @@ right now. **Any progress line you were about to type into the chat is a panel u
 **Two transports, one document.** When the tools `panel_set` / `panel_patch` / `panel_get` are
 available, this console is a pane of agent-terminal: publish the panel through them (§4.2) and
 **stop writing `.claude/work/<work>/panel.json`** — while a pane is claimed, its file is not read at
-all. When the tools are not there, write the file exactly as before (§4.3). No work folder
-(lightweight `respond`/`green`) → nothing to publish, either way.
+all. When they are not there, write the file exactly as before (§4.3). No work folder (lightweight
+`respond`/`green`) → nothing to publish, either way.
+
+**Look for them before you conclude they are missing**, once per session, in pre-flight. A harness
+namespaces an MCP server's tools, so what you are looking for is a tool whose name *ends* in
+`panel_set` — `mcp__panel__panel_set` on Claude Code, another prefix elsewhere — and the name is
+whatever it is called there, not `panel_set`. It may also be **deferred**: listed by name with no
+schema, uncallable until you load it, and indistinguishable from absent unless you read that list.
+A deferred tool is a tool that is there; load the three schemas and publish. The file is the answer
+only after that search comes back empty — and this is the one paragraph that decides it, because
+prose describing a file is otherwise obeyed to the letter.
 
 ### 4.1 The document — the same whichever transport carries it
 

package/adapters/codex/skills/flow-doctor/SKILL.md

@@ -199,7 +199,7 @@ refuse) · **degraded** (it runs, quietly worse than the config promises), plus
   `.agents/agents`, or a plugin. Report each missing name and that it falls back to `general-purpose`
   (a skill: skipped) — a panel missing agents reviews less and still reports a clean pass.
 - Harness cannot set a model per subagent → note that every `models.*` value degrades to inheritance.
-- **Where the live panel will go this session** (flow-core §4): `panel_set` / `panel_patch` exposed → the terminal draws the panel and no `panel.json` is written; absent → the panel is that file, as before. Neither is a fault — one line saying which one this session gets.
+- **Where the live panel will go this session** (flow-core §4): a tool whose name ends in `panel_set` / `panel_patch` — namespaced by the harness (`mcp__panel__panel_set` on Claude Code) and possibly **deferred**, which reads as absent until the list of deferred tools is checked → the terminal draws the panel and no `panel.json` is written; genuinely absent → the panel is that file, as before. Neither is a fault — one line naming the tools as they are actually called here, or saying the file is the transport.
 
 ### 2.3 Knowledge sources
 

package/adapters/codex/skills/flow-work-README/SKILL.md

@@ -159,7 +159,7 @@ This system **orchestrates** the project's existing sub-agents and skills (it do
 
 ## The panel document
 
-`meta.json` is the *state machine*; the panel is the *view* a reader outside the chat watches. Optional: a work that never publishes it still resolves from `meta.json` alone. Writer rules (order, when, ceiling) and the two transports — the terminal's `panel_set` / `panel_patch` / `panel_get` tools, or `.claude/work/<work>/panel.json` — are flow-core §4. The document below is the same either way.
+`meta.json` is the *state machine*; the panel is the *view* a reader outside the chat watches. Optional: a work that never publishes it still resolves from `meta.json` alone. Writer rules (order, when, ceiling) and the two transports — the terminal's `panel_set` / `panel_patch` / `panel_get` tools (namespaced by the harness, and possibly deferred: flow-core §4 says how to find them), or `.claude/work/<work>/panel.json` — are flow-core §4. The document below is the same either way.
 
 ```json
 {

package/adapters/gemini/CORE.md

@@ -16,7 +16,7 @@ Every `/flow:*` command assumes these rules. They are stated once, here, so a co
 carries what is specific to its phase. Read this once per session; a command that says "load
 `flow-core`" means this file.
 
-**This file belongs to flow `0.71.0`.** Compare it once, at the start of the session, against
+**This file belongs to flow `0.72.0`.** Compare it once, at the start of the session, against
 `version` in `~/.claude/flow/.claude-plugin/plugin.json`. The two differing means the session
 is running a **mixture** — the commands from one copy of the plugin, these shared rules from another
 — which is exactly what happens when a branch or a release candidate is loaded over an installed
@@ -268,8 +268,17 @@ right now. **Any progress line you were about to type into the chat is a panel u
 **Two transports, one document.** When the tools `panel_set` / `panel_patch` / `panel_get` are
 available, this console is a pane of agent-terminal: publish the panel through them (§4.2) and
 **stop writing `.claude/work/<work>/panel.json`** — while a pane is claimed, its file is not read at
-all. When the tools are not there, write the file exactly as before (§4.3). No work folder
-(lightweight `respond`/`green`) → nothing to publish, either way.
+all. When they are not there, write the file exactly as before (§4.3). No work folder (lightweight
+`respond`/`green`) → nothing to publish, either way.
+
+**Look for them before you conclude they are missing**, once per session, in pre-flight. A harness
+namespaces an MCP server's tools, so what you are looking for is a tool whose name *ends* in
+`panel_set` — `mcp__panel__panel_set` on Claude Code, another prefix elsewhere — and the name is
+whatever it is called there, not `panel_set`. It may also be **deferred**: listed by name with no
+schema, uncallable until you load it, and indistinguishable from absent unless you read that list.
+A deferred tool is a tool that is there; load the three schemas and publish. The file is the answer
+only after that search comes back empty — and this is the one paragraph that decides it, because
+prose describing a file is otherwise obeyed to the letter.
 
 ### 4.1 The document — the same whichever transport carries it
 

package/adapters/gemini/commands/flow/doctor.toml

@@ -195,7 +195,7 @@ refuse) · **degraded** (it runs, quietly worse than the config promises), plus
   `.agents/agents`, or a plugin. Report each missing name and that it falls back to `general-purpose`
   (a skill: skipped) — a panel missing agents reviews less and still reports a clean pass.
 - Harness cannot set a model per subagent → note that every `models.*` value degrades to inheritance.
-- **Where the live panel will go this session** (flow-core §4): `panel_set` / `panel_patch` exposed → the terminal draws the panel and no `panel.json` is written; absent → the panel is that file, as before. Neither is a fault — one line saying which one this session gets.
+- **Where the live panel will go this session** (flow-core §4): a tool whose name ends in `panel_set` / `panel_patch` — namespaced by the harness (`mcp__panel__panel_set` on Claude Code) and possibly **deferred**, which reads as absent until the list of deferred tools is checked → the terminal draws the panel and no `panel.json` is written; genuinely absent → the panel is that file, as before. Neither is a fault — one line naming the tools as they are actually called here, or saying the file is the transport.
 
 ### 2.3 Knowledge sources
 

package/adapters/gemini/commands/flow/work/README.toml

@@ -155,7 +155,7 @@ This system **orchestrates** the project's existing sub-agents and skills (it do
 
 ## The panel document
 
-`meta.json` is the *state machine*; the panel is the *view* a reader outside the chat watches. Optional: a work that never publishes it still resolves from `meta.json` alone. Writer rules (order, when, ceiling) and the two transports — the terminal's `panel_set` / `panel_patch` / `panel_get` tools, or `.claude/work/<work>/panel.json` — are flow-core §4. The document below is the same either way.
+`meta.json` is the *state machine*; the panel is the *view* a reader outside the chat watches. Optional: a work that never publishes it still resolves from `meta.json` alone. Writer rules (order, when, ceiling) and the two transports — the terminal's `panel_set` / `panel_patch` / `panel_get` tools (namespaced by the harness, and possibly deferred: flow-core §4 says how to find them), or `.claude/work/<work>/panel.json` — are flow-core §4. The document below is the same either way.
 
 ```json
 {

package/adapters/hermes/CORE.md

@@ -17,7 +17,7 @@ Every `/flow-*` command assumes these rules. They are stated once, here, so a co
 carries what is specific to its phase. Read this once per session; a command that says "load
 `flow-core`" means this file.
 
-**This file belongs to flow `0.71.0`.** Compare it once, at the start of the session, against
+**This file belongs to flow `0.72.0`.** Compare it once, at the start of the session, against
 `version` in `~/.claude/flow/.claude-plugin/plugin.json`. The two differing means the session
 is running a **mixture** — the commands from one copy of the plugin, these shared rules from another
 — which is exactly what happens when a branch or a release candidate is loaded over an installed
@@ -269,8 +269,17 @@ right now. **Any progress line you were about to type into the chat is a panel u
 **Two transports, one document.** When the tools `panel_set` / `panel_patch` / `panel_get` are
 available, this console is a pane of agent-terminal: publish the panel through them (§4.2) and
 **stop writing `.claude/work/<work>/panel.json`** — while a pane is claimed, its file is not read at
-all. When the tools are not there, write the file exactly as before (§4.3). No work folder
-(lightweight `respond`/`green`) → nothing to publish, either way.
+all. When they are not there, write the file exactly as before (§4.3). No work folder (lightweight
+`respond`/`green`) → nothing to publish, either way.
+
+**Look for them before you conclude they are missing**, once per session, in pre-flight. A harness
+namespaces an MCP server's tools, so what you are looking for is a tool whose name *ends* in
+`panel_set` — `mcp__panel__panel_set` on Claude Code, another prefix elsewhere — and the name is
+whatever it is called there, not `panel_set`. It may also be **deferred**: listed by name with no
+schema, uncallable until you load it, and indistinguishable from absent unless you read that list.
+A deferred tool is a tool that is there; load the three schemas and publish. The file is the answer
+only after that search comes back empty — and this is the one paragraph that decides it, because
+prose describing a file is otherwise obeyed to the letter.
 
 ### 4.1 The document — the same whichever transport carries it
 

package/adapters/hermes/skills/flow-doctor/SKILL.md

@@ -199,7 +199,7 @@ refuse) · **degraded** (it runs, quietly worse than the config promises), plus
   `.agents/agents`, or a plugin. Report each missing name and that it falls back to `general-purpose`
   (a skill: skipped) — a panel missing agents reviews less and still reports a clean pass.
 - Harness cannot set a model per subagent → note that every `models.*` value degrades to inheritance.
-- **Where the live panel will go this session** (flow-core §4): `panel_set` / `panel_patch` exposed → the terminal draws the panel and no `panel.json` is written; absent → the panel is that file, as before. Neither is a fault — one line saying which one this session gets.
+- **Where the live panel will go this session** (flow-core §4): a tool whose name ends in `panel_set` / `panel_patch` — namespaced by the harness (`mcp__panel__panel_set` on Claude Code) and possibly **deferred**, which reads as absent until the list of deferred tools is checked → the terminal draws the panel and no `panel.json` is written; genuinely absent → the panel is that file, as before. Neither is a fault — one line naming the tools as they are actually called here, or saying the file is the transport.
 
 ### 2.3 Knowledge sources
 

package/adapters/hermes/skills/flow-work-README/SKILL.md

@@ -159,7 +159,7 @@ This system **orchestrates** the project's existing sub-agents and skills (it do
 
 ## The panel document
 
-`meta.json` is the *state machine*; the panel is the *view* a reader outside the chat watches. Optional: a work that never publishes it still resolves from `meta.json` alone. Writer rules (order, when, ceiling) and the two transports — the terminal's `panel_set` / `panel_patch` / `panel_get` tools, or `.claude/work/<work>/panel.json` — are flow-core §4. The document below is the same either way.
+`meta.json` is the *state machine*; the panel is the *view* a reader outside the chat watches. Optional: a work that never publishes it still resolves from `meta.json` alone. Writer rules (order, when, ceiling) and the two transports — the terminal's `panel_set` / `panel_patch` / `panel_get` tools (namespaced by the harness, and possibly deferred: flow-core §4 says how to find them), or `.claude/work/<work>/panel.json` — are flow-core §4. The document below is the same either way.
 
 ```json
 {

package/adapters/opencode/CORE.md

@@ -16,7 +16,7 @@ Every `/flow-*` command assumes these rules. They are stated once, here, so a co
 carries what is specific to its phase. Read this once per session; a command that says "load
 `flow-core`" means this file.
 
-**This file belongs to flow `0.71.0`.** Compare it once, at the start of the session, against
+**This file belongs to flow `0.72.0`.** Compare it once, at the start of the session, against
 `version` in `~/.claude/flow/.claude-plugin/plugin.json`. The two differing means the session
 is running a **mixture** — the commands from one copy of the plugin, these shared rules from another
 — which is exactly what happens when a branch or a release candidate is loaded over an installed
@@ -268,8 +268,17 @@ right now. **Any progress line you were about to type into the chat is a panel u
 **Two transports, one document.** When the tools `panel_set` / `panel_patch` / `panel_get` are
 available, this console is a pane of agent-terminal: publish the panel through them (§4.2) and
 **stop writing `.claude/work/<work>/panel.json`** — while a pane is claimed, its file is not read at
-all. When the tools are not there, write the file exactly as before (§4.3). No work folder
-(lightweight `respond`/`green`) → nothing to publish, either way.
+all. When they are not there, write the file exactly as before (§4.3). No work folder (lightweight
+`respond`/`green`) → nothing to publish, either way.
+
+**Look for them before you conclude they are missing**, once per session, in pre-flight. A harness
+namespaces an MCP server's tools, so what you are looking for is a tool whose name *ends* in
+`panel_set` — `mcp__panel__panel_set` on Claude Code, another prefix elsewhere — and the name is
+whatever it is called there, not `panel_set`. It may also be **deferred**: listed by name with no
+schema, uncallable until you load it, and indistinguishable from absent unless you read that list.
+A deferred tool is a tool that is there; load the three schemas and publish. The file is the answer
+only after that search comes back empty — and this is the one paragraph that decides it, because
+prose describing a file is otherwise obeyed to the letter.
 
 ### 4.1 The document — the same whichever transport carries it
 

package/adapters/opencode/commands/flow-doctor.md

@@ -197,7 +197,7 @@ refuse) · **degraded** (it runs, quietly worse than the config promises), plus
   `.agents/agents`, or a plugin. Report each missing name and that it falls back to `general-purpose`
   (a skill: skipped) — a panel missing agents reviews less and still reports a clean pass.
 - Harness cannot set a model per subagent → note that every `models.*` value degrades to inheritance.
-- **Where the live panel will go this session** (flow-core §4): `panel_set` / `panel_patch` exposed → the terminal draws the panel and no `panel.json` is written; absent → the panel is that file, as before. Neither is a fault — one line saying which one this session gets.
+- **Where the live panel will go this session** (flow-core §4): a tool whose name ends in `panel_set` / `panel_patch` — namespaced by the harness (`mcp__panel__panel_set` on Claude Code) and possibly **deferred**, which reads as absent until the list of deferred tools is checked → the terminal draws the panel and no `panel.json` is written; genuinely absent → the panel is that file, as before. Neither is a fault — one line naming the tools as they are actually called here, or saying the file is the transport.
 
 ### 2.3 Knowledge sources
 

package/adapters/opencode/commands/flow-work-README.md

@@ -157,7 +157,7 @@ This system **orchestrates** the project's existing sub-agents and skills (it do
 
 ## The panel document
 
-`meta.json` is the *state machine*; the panel is the *view* a reader outside the chat watches. Optional: a work that never publishes it still resolves from `meta.json` alone. Writer rules (order, when, ceiling) and the two transports — the terminal's `panel_set` / `panel_patch` / `panel_get` tools, or `.claude/work/<work>/panel.json` — are flow-core §4. The document below is the same either way.
+`meta.json` is the *state machine*; the panel is the *view* a reader outside the chat watches. Optional: a work that never publishes it still resolves from `meta.json` alone. Writer rules (order, when, ceiling) and the two transports — the terminal's `panel_set` / `panel_patch` / `panel_get` tools (namespaced by the harness, and possibly deferred: flow-core §4 says how to find them), or `.claude/work/<work>/panel.json` — are flow-core §4. The document below is the same either way.
 
 ```json
 {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "flow-workflows",
-  "version": "0.71.0",
+  "version": "0.72.0",
   "description": "Guided feat/bug workflows (start→ship), post-deploy monitoring and multi-agent code review, stack-agnostic. Each repo uses a shared FLOW.md plus optional harness overlay.",
   "type": "module",
   "bin": {

package/plugins/flow/.claude-plugin/plugin.json

@@ -3,7 +3,7 @@
   "name": "flow",
   "displayName": "Flow — guided dev workflows",
   "description": "Guided feat/bug workflows (start→ship), post-deploy monitoring and multi-agent code review, stack-agnostic. Each repo uses a shared FLOW.md plus optional harness overlay.",
-  "version": "0.71.0",
+  "version": "0.72.0",
   "author": {
     "name": "Alberto Vioque",
     "email": "mashware@gmail.com"

package/plugins/flow/CHANGELOG.md

@@ -5,6 +5,21 @@ plugin and is what `/flow:news` reads to show you what changed since your previo
 
 The canonical, richest notes live in the [GitHub Releases](https://github.com/mashware/flow-workflows/releases).
 
+## v0.72.0 — The tools were there, under a name nobody looked for  ·  2026-09-22
+
+**In short**
+- **A session with the panel tools connected kept writing the file.** The rule named them `panel_set` / `panel_patch`, and a harness namespaces an MCP server's tools — so the agent looked for a name that exists nowhere and concluded, reasonably, that there was no panel.
+- **Now the search is the rule**: once per session, in pre-flight, look for a tool whose name *ends* in `panel_set` — `mcp__panel__panel_set` on Claude Code, another prefix elsewhere.
+- **A deferred tool is a tool that is there.** Listed by name with no schema, it reads as absent to anything that does not check that list; flow loads the three schemas and publishes.
+- **The file is the answer only after that search comes back empty**, and `/flow:doctor` names the tools as they are actually called here.
+
+v0.71.0 gave the panel a second transport and described it with the names the server declares.
+Neither the plugin nor the agent reading it ever sees those: a harness prefixes them, and it may
+hand them over as names alone, uncallable until their schemas are loaded. Both together are
+indistinguishable from an absent server — and the fallback, "write the file exactly as before", is
+the most obeyed sentence in the section. So the transport chosen itself was the file, every time,
+on the one terminal that had the panel.
+
 ## v0.71.0 — The panel was a file because nothing better was listening  ·  2026-09-22
 
 **In short**