From a0650c1f0ea6e0b4de82a19d024e09e5876c2358 Mon Sep 17 00:00:00 2001 From: ruv Date: Tue, 6 Oct 2026 14:42:26 -0400 Subject: [PATCH] docs: Workflows demo SVG with pinned section headers and no real peer names The SVG merged in #3847 showed the Conversation target list from the recording machine's real federation peers. Those hostnames are replaced with same-length neutral names, and the animation now pins each section's header (sticky, following the section at the top of the dock). Co-Authored-By: RuFlo --- docs/assets/ruflo-console-workflows.svg | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/assets/ruflo-console-workflows.svg b/docs/assets/ruflo-console-workflows.svg index 7a81e7c4c33..12f18d7624a 100644 --- a/docs/assets/ruflo-console-workflows.svg +++ b/docs/assets/ruflo-console-workflows.svg @@ -1 +1 @@ -ruflo console: the Workflows board (sample run)>> ruflo console: first open, the boot✕}*1▐1]*1>▐!1==▐=&~~~~~&~~~~~~~~~~1+▐0<▐+?+@{#&▐[<?@{+▐#{+/?@?0?▐?+#0/@+#0▐?0#]%~~~~~*~~~~~\#\[>+%]>▐>]*!\}=}▐\}$*\$>*>1%▐\**>> Boot: every area is checked, not just listed%$!1▐>]}1%=!$>\*]%]*▐1}1!=$}]%1>]▐%\*@&!+~~~~~{%~~~~~@=~~~~~%++]#={0&▐<@{@#/?@</#+{▐&+?/&/#+<<[&/#▐0<&<▐+?[<00?/&+▐?+?{//</<[[▐+[%&@![%!<{<~~~~~@*&/~~~~~*{%{$?~~~~~1[0**#$]$▐%1*$%=}\*=*1>▐$*1]$%%}▐\*▐<0/&[{▐0+<<[?&▐+<]+?=$00%@}~~~~~&1/!&>~~~~~0$[*[~~~~~}<=<▐\>]*\]\▐*>\$▐1▐▐@0{▐&0%#1>]]#?\<$0/~~~~~@!?1/!{\@%~~~~~!#1#!~~~~~=+[%$0▐>▐%▐%▐▐▐/}=]!>/!/}&<~~~~~#]/}?{\\/>#$~~~~~*}]~~~~~{$/>{\▐▐▐▐▐▐*+&}@+@/==~~~~~]@$&#!+$#~~~~~/{&~~~~~\{}/\*[<*▐▐▐#0$<{%%@~~~~~%00\/$?~~~~~#+1>+~~~~~*0$#\#!]{+\▐▐▐▐▐▐?>+${}@]&&~~~~1{![~~~~~+!1&!0~~~+1%}@\}/<*$▐▐▐> handLOADING []1%reads 0/14RUV.$//> >$<$<<\< /1&&$%>0//\\ &$<0 < %>%#< >1\%%<▸plugins 1▐▐▐&%@>+=[1<%+/~~~~$#~~=?~$!+$&*~~{>]&%$<}\%[▐▐▐> handshake ok·LOADING []4%reads 0/14RUV.NET // RU>$1 %1&%&$1<10 %\/> > /#/>$>%>\\#[ .. ]Missions▐▐▐!#$[>{1[1+*&1$~%~~+%*~$0%{$/{/}&=~*?0>?<&$0=> handshake ok·node demoLOADING []6%reads 0/14RUV.NET // RUVECTOR C#&###&11<\ <&& $ %<%\ #0\&%<[ OK ]Missions[ .. ]Overview]@}{\[>#\!<]@}~\+*<=@>?]0><$[!~$/[>&@%&=/!LOADING []9%reads 0/14RUV.NET // RUVECTOR CONSTELLAT#$$<&1 1 %%&/# <\<&>[ OK ]Missions[ OK ]Overview[ .. ]Swarm&1/>\@>&~0!?]@*?1+}?=~#$@~?10}\0!&1LOADING []12%reads 0/14RUV.NET // RUVECTOR CONSTELLATION▐r$$>&%#0$>#\[ OK ]Missions[ OK ]Overview[ OK ]Swarm[ OK ]Workflows[ .. ]Hive-Mind#]][~&]0>&$+*~~/%~~~#~>#]!{\0LOADING []14%reads 0/14RUV.NET // RUVECTOR CONSTELLATION▐rUv·rufl&\$&%1[ OK ]Workflows[ OK ]Hive-Mind[ .. ]Claims▐▐▐/~?~~>~~~~~~~~#~$▐▐▐LOADING []19%reads 0/14RUV.NET // RUVECTOR CONSTELLATION▐rUv·ruflo v0.34.1[ OK ]Workflows[ OK ]Hive-Mind[ OK ]Claims[ OK ]Approvals[ .. ]Automation▐▐▐~~~~~~~~~~~~~~~▐▐▐LOADING []21%reads 0/14[ OK ]Approvals[ OK ]Automation[ .. ]Learning~~~~~~~~~~~~~~~LOADING []24%reads 0/1401001111 00111101 11111000[ OK ]Approvals[ OK ]Automation[ OK ]Learning[ OK ]Neural[ .. ]Vector Lab~~~~~~~~~~~~~~~LOADING []27%reads 0/1401110011 11001111 11111110[ OK ]Neural[ OK ]Vector Lab[ .. ]Memory Lab▐▐▐~~~~~~~~~~~~~~~▐▐▐LOADING []29%reads 0/1401110010 11100111 01111111[ OK ]Neural[ OK ]Vector Lab[ OK ]Memory Lab[ .. ]Harness~~~~~~~~~~~~~~~LOADING []31%reads 0/1401110010 01100111 01111111[ OK ]Harness[ .. ]Evolution%~*~~~~~~~~~~~~~LOADING []34%reads 0/1401110010 01010001 11011111[ OK ]Harness[ OK ]Evolution[ OK ]Security[ .. ]Federation~~~~~~~~~~~~~~~LOADING []36%reads 0/1401110010 01010101 11101111[ OK ]Federation[ .. ]x.ruv.io~~~~~~~~~~~~~~~LOADING []39%reads 0/1401110010 01010101 01110111[ OK ]Federation[ OK ]x.ruv.io[ .. ]Plugins~~~~~~~~~~~~~~~LOADING []41%reads 0/1401110010 01010101 01110011[ OK ]Federation[ OK ]x.ruv.io[ OK ]Plugins[ .. ]Skills~~~~~~~~~~~~~~~LOADING []44%reads 0/1401110010 01010101 01110110=rUv[ OK ]Skills[ OK ]Catalog[ .. ]Dev Tools~~~~~~~~~~~~~~~LOADING []48%reads 0/14[ OK ]Skills[ OK ]Catalog[ OK ]Dev Tools[ .. ]Sandbox~~~~~~~~~~~~~~~LOADING []50%reads 0/14[ OK ]Sandbox[ .. ]Cost~~~~~~~~~~~~~~~LOADING []53%reads 0/14[ OK ]Sandbox[ OK ]Cost[ OK ]Timeline[ .. ]Events▐▐▐~~~~~~~~~~~~~~~▐▐▐LOADING []55%reads 0/14[ OK ]Events[ .. ]The Room~~~~~~~~~~~~~~~LOADING []58%reads 0/14[ OK ]Events[ OK ]The Room[ .. ]Perf~~~~~~~~~~LOADING []61%reads 0/14[ OK ]Events[ OK ]The Room[ OK ]Perf[ OK ]Terminal[ .. ]Settings~~~~~LOADING []63%reads 0/14[ OK ]Terminal[ OK ]Settings~~~~~~~~~~~~~~~LOADING []66%reads 0/14[WARN] READY·29/29 verified·9 stars dark·press a key>> the menu enters: banner, then each card~~~~~~~~~~~~~~~LOADING []67%reads 0/14LOADING []69%reads 0/14~~~~~~~~~~~~~~~LOADING []70%reads 0/14~~~~~~~~~~~~~~~~~~~~LOADING []70%reads 0/14▐▐▐~~~~~▐??▐▐#⌂⌘ʔ⛭⟳RUFLOx.ruv.ioAGENTS WELCOME.NETWORKS:x.ruv.io*relay.ruv.io*agentbbs*mcp*claude code[0:📟MENU]SWARMMINDSAFETYNETWORKTOOLS🔎:find a pageMISSION CONTROL✎goal:what do you want done? e.g. add a dark mode toggle to se…research→create (ADRs, SOP)→build→test→validate→secure→benchmark→learn(1) open Missionsgroups[▾expand all ][▸collapse all ]▾▾⌂⌘ʔ⛭⟳✕⌂⌘ʔ⛭⟳RUFLOx.ruv.ioAGENTS WELCOME.NETWORKS:x.ruv.io*relay.ruv.io*agentbbs*mcp*claude code[0:📟MENU]SWARMMINDSAFETYNETWORKTOOLS🔎:find a pageMISSION CONTROL✎goal:what do you want done? e.g. add a dark mode toggle to se…research→create (ADRs, SOP)→build→test→validate→secure→benchmark→learn(1) open Missionsgroups[▾expand all ][▸collapse all ]▾▾⌂⌘ʔ⛭⟳>>🔀:: Claude Code workflow runs and the ruflo sw…WORKFLOWS0:📟MAIN[SWARM]MINDSAFETYNETWORKTOOLS1:🎯Missions2:🏠Overview3:🐝Swarm[🔀WORKFLOWS]b:👑Hive-Mind4:📌Claimsq:✅Approvals[LINK OK]▸sync…keys on[ Palette ][ x ][✦][ r ][ h ][×]WORKFLOWSreading workflow runs…read when this page opens, then on each refresh while it stays open⌂⌘ʔ⛭⟳WORKFLOWS2 runs·j/k move·b/l column·d inspectsample: build-tune-reviewrunning8 agents·2 running·6 done·0 failed·312.0k tok·6m59s▸1 sample: build-tun…2 sample: build-tun…[ ] switch runPHASESTune·latency, cache and memory passes✔1 Build3/3✔tune-latency52.1k tok2m20s▸❯◓2 Tune2/3✔tune-cache9.8k tok45s◓3 Review1/2◓tune-memory44.7k tok2m24sFigures are derived from the journal and transcripts (read up to 3…le-btr·d inspects the agent·u/i switches run[ prev ][ next ][◂phases ][ agents▸][◂run ][ run▸][ Inspect ][ worktrees ][ next template ][▸deeper ][◂back ][▾next ][▴prev ]Stop, message and redirect: the control tab calls the engine's own…MISSION AUTOPILOTautopilot is offidle·wired·0 running·0 done (0 unverified)·0 failed·rea…[ Stop (9) ][ Check now ]Start is the action row (key 8) and asks first·Stop asks nothing⌂⌘ʔ⛭⟳%⌂⌘ʔ⛭⟳=⌂⌘ʔ⛭⟳>⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳>> Workflows: the run board, sample run, live⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳8 agents·2 running·6 done·0 failed·312.0k tok·7m01s◓3 Review1/2◓tune-memory44.7k tok2m26s>> Phases: j/k move, b/l pick the column⌂⌘ʔ⛭⟳PHASESReview·security and style reviewers✔1 Build3/3◓review-security71.6k tok1m31s❯◓2 Tune2/3✔review-style21.2k tok35s▸◓3 Review1/2⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳@⌂⌘ʔ⛭⟳✔1 Build3/3▸✔tune-latency52.1k tok2m20s❯◐2 Tune2/3✔tune-cache9.8k tok45s◐3 Review1/2◐tune-memory44.7k tok2m26s⌂⌘ʔ⛭⟳INSPECTtune-latencyagenta4sample0004·phase Tune·donemodelSonnet 5.5 (claude-sonnet-5-5)tokens52.1k context tokens of its latest requestelapsed2m20s·started 16:43:04Zworktree/work/demo/.worktrees/tune-latencytool calls4·last Bashresult: p95 latency 212 ms -> 148 ms by batching the two sequentia…transcript/tmp/demo-cfg/cfg.INcd/projects/-tmp-demo/5a3c…[ Open transcript ]⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳8 agents·2 running·6 done·0 failed·312.0k tok·7m04s❯◓2 Tune2/3✔tune-cache9.8k tok45s◓3 Review1/2◓tune-memory44.7k tok2m29s⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳>> Inspect: model, worktree, tokens, time⌂⌘ʔ⛭⟳❯◒2 Tune2/3▸✔tune-cache9.8k tok45s◒3 Review1/2◒tune-memory44.7k tok2m29sINSPECTtune-cacheagenta5sample0005·phase Tune·donemodelHaiku 4.5 (claude-haiku-4-5)tokens9.8k context tokens of its latest requestelapsed45s·started 16:43:09Zworktreenonetool calls2·last Bashresult: No changes made❯◒2 Tune2/3✔tune-cache9.8k tok45s◒3 Review1/2▸◒tune-memory44.7k tok2m29sINSPECTtune-memoryagenta6sample0006·phase Tune·runningtokens44.7k context tokens of its latest requestelapsed2m29s·started 16:45:39Zworktree/work/demo/.worktrees/tune-memorytool calls5·last BashStop, message and redirect: the control tab calls the engine's own…⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳◓3 Review1/2▸◓tune-memory44.7k tok2m29s8 agents·2 running·6 done·0 failed·312.0k tok·7m07s◓3 Review1/2▸◓tune-memory44.7k tok2m32selapsed2m32s·started 16:45:39Z❯◑2 Tune2/3✔tune-cache9.8k tok45s◑3 Review1/2▸◑tune-memory44.7k tok2m32s[ prev ][ next ][◂phases ][ agents▸][◂run ][ run▸][ Closetabs[●detail ][○protector ][○mission ][○guide ][○control ]Anatolenot installedpermissionsyour settings allow every classspendhour n/a/n/a·day n/a/n/a·total n/a/n/a (le…journal0 kB of 1500 kB capno envelope approved yet: set one below and press Startno adaptation yet: it needs 6 or more verified steps on a setting…next envelope (a change asks again):classes[●read ][●edit ][●test ][●git-local ][○git-branch ][○network ][○spawn ][○mcp ]hour[−] $2 [+]day[−] $10 [+]total[−] $40 [+]at once[−] 1 [+]duration[−] 7 d [+][○accept running without Anatole ][ load draft file ].claude-flow/console/autopilot/envelope.draft.json holds paths, r…draft is valid·hash ae2ec683d57d·/tmp/demoInside the envelope autopilot does not ask this console. It never…It is not a sandbox (the envelope gates what a task says) and not…AUTOPILOT, PARKED FOR YOUautopilot has not started: nothing is parked>> Control tab: Stop asks first (then we cancel)proof: verified·TaskStop worked on a real background agent (t…[ stop the whole run sample: build-tune-review ]call: TaskStop {"task_id":"wf_sample-btr"}proof: unverified·TaskStop with a workflow run's id was not t…[ message the agent ]call: none (type the guidance first (the field above))proof: queued-only·SendMessage queued a message for a running…[ redirect: stop, then resume with guidance ]proof: unverified·Half one is the stop (see above); half two…A refusal by the engine is shown and nothing else is tried in its…[ stop agent tune-memory ]call: TaskStop {"task_id":"a6sample0006"}tabs[○detail ][○protector ][○mission ][○guide ][●control ]runsample: build-tune-reviewagenttune-memorytool callsbound: each call goes through the engine's per…message:what to tell the agent, or what to change on a redirectmessage: none typed yet▶CONFIRM NEEDEDclick Yes or press yConfirm: stop agent tune-memory?runs: TaskStop {"task_id":"a6sample0006"}Effect: Calls the engine's TaskStop through the session's ownpermission check and dialog. Proven on a background Agent-toolagent; for an agent of a workflow run the engine's answer is theproof. Stopped work is not undone by resuming: a resumed runstarts its unfinished agents again.[ Yes, run it (y) ][ Cancel (n) ]elapsed2m41s·started 16:45:39Zelapsed2m44s·started 16:45:39Z>> Mission autopilot: envelope draft, not startedDRILL-DOWN[●network ][○spawn ][○mcp ]draft is valid·hash 6fbfd921accf·/tmp/demo>> Kill switch: Stop (9) is always there; Start asks[ Drill into this run ]e deeper·s search—runs›phases›agen…TRIAGE1 failed·1 empty·4 ok·2 running or waiting✖review-style·Review·error·Error: lint config .eslintrc.jso…○tune-cache·Tune·empty·its whole result is "No changes made"re-run just the failed items (this text is for you to run; theconsole does not):Stop the run first (the Control tab's Stop asks first and callsTaskStop; unverified for a workflow run's own id: Claude Code'sWorkflows panel always works), then:Workflow({ scriptPath: "<the script that started this run>",resumeFromRunId: "wf_sample-btr" })stays cached, its error is its recorded result: review-style(Review). Change that agent() call or its input to run it againClasses come from the run record, the journal's results and thetranscripts; no timeout event is written, so timeout is a textmatch.SEARCHsearch:2+ characters: a run, label, tool call, transcript text or m…Type to search every level at once; Enter jumps to the first hit.>> Parked queue: empty, the autopilot never startedCOSTpricesruflo-cost-tracker book as of 2026-10-03·USD…run$1.00·859.8k billed tokens (all request buck…>> Drill-down: run > phase > agent[ Runs: sample: build-t…]›[ Phases: Tune ]›Agents: tune-memory[▸deeper ][◂back ][▴prev ][▾next ]tune-latencydone·4 calls·2m20stune-cachedone·2 calls·45s▸tune-memoryrunning·5 calls·2m53skeys: e deeper·z back·m next·a prev·q next tab·f follow…[ Runs: sample: build-t…][ Phases: Tune ]…[ Agents: tune-memory ]Agent·Activityview[●Activity ][○Log ][○Files ][○Result ]tune-memory·running·44.7k tok·claude-sonnet-5-5transcript: whole file, 4 KB of the 3.0 MB read cap·12 entries·…▸#1✔Bashnode bench/memory.js --baseli…+0ms900ms#2✔Readsrc/pool.ts+21.8s900ms#3✔Grepnew Array+43.7s900ms#4✔Editsrc/pool.ts+65.5s900ms#5✔Bashnode bench/memory.js+87.3s900ms0 later·call 1 of 5·open it for the full input and output>> Agent tabs: Activity, Log, Files, ResultLogview[○Activity ][●Log ][○Files ][○Result ][●following ][ level: all lines ][ scope: this agent ]16:45:39userSAMPLE RUN—tune-memory: tune step of the d…16:45:50assiStep 1: Bash on the demo project.16:45:50toolBash node bench/memory.js --baseline16:46:12assiStep 2: Read on the demo project.16:46:12toolRead src/pool.ts16:46:34assiStep 3: Grep on the demo project.16:46:34toolGrep new Array16:46:56assiStep 4: Edit on the demo project.16:46:56toolEdit src/pool.ts16:47:18assiStep 5: Bash on the demo project.16:47:18toolBash node bench/memory.js▸16:47:50assiWorking on it.0 above·0 below·line 12 of 12follow pins the newest line and is re-drawn on this page's refresh…Filesview[○Activity ][○Log ][●Files ][○Result ]worktree /work/demo/.worktre…[ Read diff stat ]diff stat not read: the worktree is not inside this project's fold…src/pool.tsedit×1 read×1Resultview[○Activity ][○Log ][○Files ][●Result ]The agent's return is not in memory (the run folder is outside Cla…[ Read the journal (up to 1.0 MB) ][ Runs: sample: build-t…][ Phases: Tune ][ Agents: tune-memory ][[▸deeper ][◂back ][▴prev ][▾next ]Agentool call Bash·#3 of 12·16:45:50·ok·900ms·input·50 characters·lines 1–3 of 3Acti{ity"command": "node bench/memory.js --baseline"}output·61 characters·lines 1–3 of 3rss 171 MBDEMO_API_‹masked›heap: pool=4096[▴up ][▾down ]6 lines in all>> A call and its output: secrets are masked▸tune-memoryrunning·5 calls·3m02s▸tune-memoryrunning·5 calls·3m05sTranscripts: up to 3.0 MB each (a larger one is read from itsend:≥); at most 24.0 MB parsed per refresh; 8 of 16 read.≥marks a floor,≈a family price.>> Triage: failed, empty, and the re-run textBuild$0.39306.1k tok▸Tune$0.28332.4k tokReview$0.32221.4k toktune-latencySonnet 5.5$0.14156.3k toktune-cacheHaiku 4.5$0.00919.6k toktune-memorySonnet 5.5$0.14156.4k tokguardsoff: set wfBudgetRunUsd, wfBudgetDayUsd or wfA…MISSION LINKS0 agent-to-task links·16 agents unlinked·0 mission tasks knownNo mission yet in the ledger or the observation: make one in Missi…agents unlinked: its label has no [mission:<id> task:<id>] tag (a…WORKTREES (READ-ONLY LIST; ONE CONFIRM-GATED REMOVAL)3 worktrees·read 46s ago·origin/main as last fetched 3d ago (…….worktrees/…↑0↓0·clean·1m…kept: outside .claude/worktrees an…>> Search: one term, every level at once….worktrees/…↑1↓0·clean·1m…kept: outside .claude/worktrees an…search:2+ characters: a run, label, tool call, transcript text…⏎go3 worktrees·read 47s ago·origin/main as last fetched 3d ago (…search:pool⏎go6 matches·Enter in the field goes to the firstAgents (1)[▸tune-memorysample: build-tune-revie…]Tool calls (2)[▸Read src/pool.tstune-memory·call #2·…][▸Edit src/pool.tstune-memory·call #4·…]Transcript text (3)[▸rss 171 MB DEMO_API_‹masked›…tune-memory·output of…][▸class Pool { /* 96 lines */ }tune-memory·output of…][▸src/pool.ts:41: new Array(40…tune-memory·output of…]looked through 8 agents' transcripts (3,435 characters of 8,000,00…8 agents' transcripts are not in memory, so they were not searched[ Read this run's transcripts (up to 24 MB in all) ]3 worktrees·read 51s ago·origin/main as last fetched 3d ago (……tmp/demo…↑0↓0·clean·1m·not att…kept: the main wo…live processes: not checked (the read button, w, checks them; the…3 worktrees·read 52s ago·origin/main as last fetched 3d ago (…cleanup: live processes not checked: press the read button firstremoved: the directory only, never the branch, never with --force…[ read worktrees + live processes ]>> Cost: per run, phase, agent (a price list)3 worktrees·read 53s ago·origin/main as last fetched 3d ago (…3 worktrees·read 54s ago·origin/main as last fetched 3d ago (…SWARM NESTINGPick the ruflo swarm run ([ ] to switch) to see how its swarm, hiv…REPLAY[▸open ]14 moments in 6m10s: scrub back and forth, jump by phase…WORKFLOW TEMPLATES (DRY RUN, THEN A CONFIRMED LAUNCH)te…[●Review ][○Migrate ][○Research ][○Build, tune, review ]Independent reviewers read the same change through one lens each, a3 worktrees·read 56s ago·origin/main as last fetched 3d ago (…verifier challenges what they found, one agent writes it up.what to review: the current uncommitted diffreviewers[−] 4 [+](2 to 8)>> Mission links: no mission yet, so none linked3 worktrees·read 57s ago·origin/main as last fetched 3d ago (…lens…[●correctness ][○security ][○performance ][○all ]3 worktrees·read 58s ago·origin/main as last fetched 3d ago (…dry run: 7 agents in 4 phases (Gather 1·Review 4·Verify 1·R…counts only: tokens, time and cost are not estimated, nothing has…launch (the action row) sends this as a visible prompt after you…3 worktrees·read 59s ago·origin/main as last fetched 3d ago (…>> Worktrees: read-only list (sample worktrees)3 worktrees·read 1m ago·origin/main as last fetched 3d ago (n…[ reading…]3 worktrees·read 1s ago·origin/main as last fetched 3d ago (n……tmp/demo…↑0↓0·1 changed·1m·3 live·…kept: the main…live processes: 686 working directories seen, 1s ago, your own pr…cleanup: nothing is both merged and cleanRUN A MISSION AS A WORKFLOWNo active mission: create one in Mission Control, and its SPARC pl…COMPARE RUNS[▸open ]1 other run of "sample: build-tune-review": agents, toke…PROJECT ANATOLE, PER AGENT (REPORTED BY THE MOD)Project Anatole is not installed or has written nothing here: cla…>> Swarm nesting: needs a ruflo swarm runEXPORT RUN SUMMARY[▸open ].claude-flow/console/exports/sample-build-tune-review-wf…>> Replay: step through the finished sample run3 worktrees·read 7s ago·origin/main as last fetched 3d ago (n…live processes: 686 working directories seen, 7s ago, your own pr…3 worktrees·read 8s ago·origin/main as last fetched 3d ago (n…▸tune-memoryrunning·5 calls·3m32sidle·not started·0 running·0 done (0 unverified)·0 failed…8 agents·2 running·6 done·0 failed·312.0k tok·8m10s◐3 Review1/2▸◐tune-memory44.7k tok3m35s◓3 Review1/2▸◓tune-memory44.7k tok3m35s⌂⌘ʔ⛭⟳[ Runs: sample: build-t…]›[ Phases: Build ]›Agents: api-core▸api-coredone·9 calls·2m30scli-flagsdone·6 calls·1m30sdocs-passdone·4 calls·1m00s1 empty·7 okrunn/a·n/a billed tokens (all request buckets)…▸Buildn/an/aTunen/an/aReviewn/an/aapi-coreOpus 5.5n/an/acli-flagsSonnet 5.5n/an/adocs-passHaiku 4.5n/an/a3 worktrees·read 18s ago·origin/main as last fetched 3d ago (…3 worktrees·read 19s ago·origin/main as last fetched 3d ago (…live processes: 686 working directories seen, 19s ago, your own p…[▸open ]16 moments in 9m20s: scrub back and forth, jump by phase…live processes: 686 working directories seen, 20s ago, your own p…[▾close ]16 moments in 9m20s: scrub back and forth, jump by phas…step 0/16·+0s of +9m20s·16x·⏸paused▸before the first agent started[⏮][◂◂phase ][◂step ][▶play ][ step▸][ phase▸▸][⏭]speed[−]16x[ + ]of 1·4·16·64·256○Build0/3○api-corequeuedn/an/a○cli-flagsqueuedn/an/a○docs-passqueuedn/an/a○Tune0/3○tune-latencyqueuedn/an/a○tune-cachequeuedn/an/a○tune-memoryqueuedn/an/a○Review0/2○review-securityqueuedn/an/a○review-stylequeuedn/an/aTokens show n/a until an agent ends: only its final figure is rec…This replay cannot rewind a run; stop or message one from its con…step 1/16·+0s of +9m20s·16x·⏸paused▸api-core started (Build)◐Build0/3·1 running◐api-corerunningn/a0sCONVERSATION[▸open ]talk to Claude, its agents, the hive, federation, Codex…>> Templates: dry run counts, nothing launchedte…[○Review ][○Migrate ][○Research ][●Build, tune, review ]Builders each write one part in their own worktree, tuners measureand improve it, two reviewers read the result.what to build:the feature and where it goes (required)builders[−] 3 [+](1 to 6)tuners[−] 1 [+](1 to 4)dry run: 6 agents in 3 phases (Build 3·Tune 1·Review 2), at m…needs: what to buildSAVED VIEWS AND PINS>> Run a mission as a workflow: no mission, so empty[▸open ]0 pinned·the cursor is remembered across restartsread 1s ago·re-read on each refresh while this page is open, nev…>> Compare runs: this sample run vs yesterday'sread 3s ago·re-read on each refresh while this page is open, nev…[▾close ]1 other run of "sample: build-tune-review": agents, tok…A prevvsB le-btr[ changed only ]stateA completedB runningagentsA 8 (8 ok, 0 failed) B 8 (6 ok, 0 failed)tokensA 374.5kB 312.0k-17%timeA 9m25sB 8m31s-9%1 same·7 changed·0 only in B (+)·0 only in A (-)~ = toke…~ api-coreA✔70.2k 2m30sB✔61.4k 1m50s…= cli-flagsA✔41.8k 1m30sB✔38.9k 1m15s…~ docs-passA✔15.1k 1m00sB✔12.3k 55s…~ tune-latencyA✔63.9k 2m50sB✔52.1k 2m20s…~ tune-cacheA✔11.4k 50sB✔9.8k 45s…~ tune-memoryA✔58.3k 3m50sB◐44.7k 3m56s…~ review-securityA✔80.8k 2m20sB◐71.6k 3m01s…~ review-styleA✔33.0k 1m20sB✔21.2k 35s…run B is derived from journal and transcripts, not a finished run…result text differs between the two (it always can; only the agen…timeA 9m25sB 8m34s-9%~ tune-memoryA✔58.3k 3m50sB◐44.7k 3m59s…~ review-securityA✔80.8k 2m20sB◐71.6k 3m04s…read 2s ago·re-read on each refresh while this page is open, nev…>> Export run summary: default path (not exported)read 0s ago·re-read on each refresh while this page is open, nev…[▾close ].claude-flow/console/exports/sample-build-tune-review-w…Writes a markdown summary: phases, agents, tokens, time, results…Allowed: under /tmp/demo·name ends in .md·no ..·never overw…path:.claude-flow/console/exports/sample-build-tune-review-wf_samp…[ Export to the default path ]>> Conversation: targets and composer, nothing sent[▾close ]talk to Claude, its agents, the hive, federation, Codex…[●@claude ]Claude (this session)·prompt·leaves: machine·se…sends: nothing beyond what a turn of this session sends to th…cost: a visible turn of this session, billed as any turn isreply: its answer is the next turn in the transcript, not cap…[○@tune-memory ]agent tune-memory·send-message·leaves: machi…[○@review-security ]agent review-security·send-message·leave…[○@task ]ruflo task note·task·leaves: machine·free[○@peer-ruvzen ]peer ruvzen·peer·leaves: tailnet·peer-sess…[○@peer-zenbook ]peer zenbook·peer·leaves: tailnet·peer-se…[○@peer-ruv-mac-mini ]peer ruv-mac-mini·peer·leaves: tailnet…[○@peer-cognitum-v0 ]peer cognitum-v0·peer·leaves: tailnet·…[○@peer-cognitum-v1 ]peer cognitum-v1·peer·leaves: tailnet·…[○@peer-cognitum-cluster-1 ]peer cognitum-cluster-1·peer·lea…[○@peer-cognitum-cluster-2 ]peer cognitum-cluster-2·peer·lea…[○@peer-cognitum-cluster-3 ]peer cognitum-cluster-3·peer·lea…[○@peer-heximal-v0 ]peer heximal-v0·peer·leaves: tailnet·p…[○@peer-heximal-cluster-1 ]peer heximal-cluster-1·peer·leave…[○@peer-heximal-cluster-2 ]peer heximal-cluster-2·peer·leave…[○@peer-heximal-cluster-3 ]peer heximal-cluster-3·peer·leave…[○@peer-pi5-ai02 ]peer Pi5-AI02·peer·leaves: tailnet·peer-…[○@peer-seed-1 ]peer seed-1·peer·leaves: tailnet·peer-sess…[○@peer-seed-2 ]peer seed-2·peer·leaves: tailnet·peer-sess…[○@peer-seed-3 ]peer seed-3·peer·leaves: tailnet·peer-sess…[○@peer-seed-4 ]peer seed-4·peer·leaves: tailnet·peer-sess…[○@peer-seed-5 ]peer seed-5·peer·leaves: tailnet·peer-sess…[○@peer-ruos-macmini-2012 ]peer ruos-macmini-2012·peer·leave…[○@codex ]Codex (codex exec, read-only)·codex·leaves: intern…[○@openrouter ]openrouter (openrouter/auto)·http·leaves: int…message:@claude @openrouter what do you think? (@all = agents, hive…>> Saved views and pins: the cursor is remembered[▾close ]0 pinned·the cursor is remembered across restartsFile .claude-flow/console/wf-views.json (≤64 kB, versioned)·no…Remembers the cursor (run, phase, agent by name), filters and sea…search: none·filters: none[ Pin this run ][ Forget saved views ]No pinned runs (up to 20).….worktrees/…↑1↓0·clean·2m…kept: outside .claude/worktrees an……tmp/demo…↑0↓0·1 changed·2m·3 live·…kept: the main…live processes: 686 working directories seen, 57s ago, your own p…….worktrees/…↑0↓0·clean·2m…kept: outside .claude/worktrees an…live processes: 686 working directories seen, 58s ago, your own p…sample: build-tune-reviewcompleted8 agents·0 running·8 done·0 failed·374.5k tok·9m25s1 sample: build-tun…▸2 sample: build-tun…[ ] switch runPHASESBuild·client, CLI flags, docs in parallel▸✔1 Build3/3✔api-core70.2k tok2m30s✔2 Tune3/3✔cli-flags41.8k tok1m30s✔3 Review2/2✔docs-pass15.1k tok1m00sFigures are from the run record Claude Code wrote when it finished.prev·d inspects the agent·u/i switches run[ Runs: sample: build-t…]›[ Phases: Tune ]›Agents: tune-latency▸tune-latencydone·4 calls·2m20stune-memoryrunning·5 calls·4m26stune-memoryrunning·5 calls·4m29s>> sample run, not real data>> ruflo console: the Workflows board (sample run)R U F L OC O N S O L Ev0.34.1the WORKFLOWS boardphases◂▸agents·drill-down·searchtriage·cost·control·autopilotevery figure comes from the run files:nothing is estimatedSAMPLE RUN: authored for this demo/plugin install ruflo-console@ruflothen/ruflo workflows +ruflo console: the Workflows board (sample run)>> ruflo console: first open, the bootBOOT✕\}]*▐\=}!\}=*1▐]=}}%!=!▐\%$<+~~~~~+~~~~~~~~~~]&!+}0<[▐@</&/?[▐?+{+{?@@{@▐{00>> Boot: every area is checked, not just listed%$!1▐>]}1%=!$>\*]%]*▐1}1!=$}]%1>]▐%\*@&!+~~~~~{%~~~~~@=~~~~~%++]#={0&▐<@{@#/?@</#+{▐&+?/&/#+<<[&/#▐0<&={0&▐<@{@#/?@</#+{+▐&+?/&/#+<<[&/#▐0<&>]*▐%$}1>*]%=!▐1}1\**$}$>1%▐>![]!+*/=~~~~~>0]%~~~~~[\[$<\~~~~~#>!//$<\{?▐@#[{[<+&[#+</▐?/&++<#/#@+▐#+▐]}\}]*\▐>}1\>1\▐\!=<${<*!*/~~~~~=<!+1~~~~~#%{%0}~~~~~/=={%▐#+<{0?[{<<▐@[+[&<#▐<▐>1$*▐=}%1▐=>0$&%01\?$?}~~~~~!}@]<*$&}~~~~~=0]<\0~~~~~{%/*<}▐/0@▐&0▐/▐▐▐&*/?{[@%+>0]1~~~~~=<%0$<<}/]?~~~~~%[%0#~~~~~\<}@]▐▐▐▐▐▐#1*!<$]{]{][0~~~~~0$[}*{=/>~~~~~=}~~~~~+*&]@%?▐▐▐0%][**!&{=~~~~~&?$\[*{1~~~~~>1$~~~~~#=[}?+/%\/▐▐▐*?\]0>?1[}~~~~~/}*!&]~~~~~]&[>~~~~~*#{$[{\*&/▐▐▐>LOADING []0%reads 0/14<$<00$1 \# $&%><\% $%>$>/&>1%< $\1< ><#>>>00▸plugins 1▐▐▐@%#?$@$0!?\=~~~~#@*~~~~~@&]/!~~~>{/%?/>\[?▐▐▐> handshakeLOADING []2%reads 0/14RUV.NET 1& /\>1>&\\>#0\<110// 00$ $ #><## %&%1\%▐▐▐$+}{]/}?%?\0}*~~~~{=]~~?]+<@}&$+*~\/?1[&@@>{>▐▐▐> handshake ok·nodeLOADING []5%reads 0/14RUV.NET // RUVECT%/ >&\&>$##> &&<\ \ 01>$1><0&$0[ .. ]Missions▐▐▐&}/]<![1[$<*0~&}~\??%[$#!/\*$&}[~<>&>/$%+\?0> handshake ok·node demoLOADING []7%reads 0/14RUV.NET // RUVECTOR CONST%%0#0\0> \# > >#&0$&>%>[ OK ]Missions[ OK ]Overview[ .. ]Swarm[1?%0=&*@={!/${![1{}01&%01?%@1\&*/&%[\<LOADING []11%reads 0/14RUV.NET // RUVECTOR CONSTELLATION #01# $><% 1>/&#/[ OK ]Missions[ OK ]Overview[ OK ]Swarm[ .. ]Workflows[$<<*~1{}@1{}[={!@~*@]~*/$?[]&>0LOADING []13%reads 0/14RUV.NET // RUVECTOR CONSTELLATION▐rUv·>\&<#>\$/[ OK ]Workflows[ .. ]Hive-Mind>?~/*{$0#$~~~~~~0~$@>>{LOADING []16%reads 0/14RUV.NET // RUVECTOR CONSTELLATION▐rUv·ruflo v0.<1&[ OK ]Workflows[ OK ]Hive-Mind[ OK ]Claims[ .. ]Approvals▐▐▐{*~1~~@=~~~~~~~~1~#!▐▐▐LOADING []19%reads 0/14RUV.NET // RUVECTOR CONSTELLATION▐rUv·ruflo v0.34.1[ OK ]Approvals[ .. ]Automation~~~~~~~~~~~~~~~LOADING []22%reads 0/1410011111 01111011 11110001[ OK ]Approvals[ OK ]Automation[ OK ]Learning[ .. ]Neural~~~~~~~~~~~~~~~LOADING []24%reads 0/1401001111 00111101 11111000[ OK ]Neural[ .. ]Vector Lab~~~~~~~~~~~~~~~LOADING []26%reads 0/1401110111 10011110 11111100[ OK ]Neural[ OK ]Vector Lab[ .. ]Memory Lab~~~~~~~~~~~~~~~LOADING []29%reads 0/1401110010 11001111 11111110[ OK ]Neural[ OK ]Vector Lab[ OK ]Memory Lab[ .. ]Harness~~~~~~~~~~~~~~~LOADING []31%reads 0/1401110010 01100111 01111111[ OK ]Harness[ OK ]Evolution[ .. ]Security%~*~~~~~~~~~~~~~LOADING []34%reads 0/1401110010 01010011 10111111[ OK ]Harness[ OK ]Evolution[ OK ]Security[ .. ]Federation~~~~~~~~~~~~~~~LOADING []37%reads 0/1401110010 01010101 11101111[ OK ]Federation[ .. ]x.ruv.io~~~~~~~~~~~~~~~LOADING []39%reads 0/1401110010 01010101 01110111[ OK ]Federation[ OK ]x.ruv.io[ .. ]Plugins~~~~~~~~~~~~~~~LOADING []43%reads 0/14[ OK ]Federation[ OK ]x.ruv.io[ OK ]Plugins[ OK ]Skills[ .. ]Catalog~~~~~~~~~~LOADING []46%reads 0/1401110010 01010101 01110110=rUv[ OK ]Skills[ OK ]Catalog[ .. ]Dev Tools~~~~~~~~~~~~~~~LOADING []48%reads 0/14[ OK ]Skills[ OK ]Catalog[ OK ]Dev Tools[ .. ]Sandbox~~~~~~~~~~~~~~~LOADING []51%reads 0/14[ OK ]Sandbox[ OK ]Cost[ .. ]Timeline~~~~~LOADING []53%reads 0/14[ OK ]Sandbox[ OK ]Cost[ OK ]Timeline[ .. ]Events~~~~~LOADING []56%reads 0/14[ OK ]Events[ .. ]The RoomLOADING []58%reads 0/14[ OK ]Events[ OK ]The Room[ OK ]Perf[ .. ]Terminal~~~~~LOADING []61%reads 0/14[ OK ]Terminal[ .. ]SettingsLOADING []64%reads 0/14[ OK ]Terminal[ OK ]SettingsLOADING []66%reads 0/14[WARN] READY·29/29 verified·9 stars dark·press a key>> the menu enters: banner, then each cardLOADING []68%reads 0/14▐▐▐~~~~~~~~~~▐▐▐LOADING []69%reads 0/14LOADING []70%reads 0/14~~~~~LOADING []70%reads 0/14⌂⌘ʔ⛭⟳RUFLOx.ruv.ioAGENTS WELCOME.NETWORKS:x.ruv.io*relay.ruv.io*agentbbs*mcp*claude code[0:📟MENU]SWARMMINDSAFETYNETWORKTOOLS🔎:find a pageMISSION CONTROL✎goal:what do you want done? e.g. add a dark mode toggle to se…research→create (ADRs, SOP)→build→test→validate→secure→benchmark→learn(1) open Missionsgroups[▾expand all ][▸collapse all ]▾▾%⌂⌘ʔ⛭⟳✕⌂⌘ʔ⛭⟳RUFLOx.ruv.ioAGENTS WELCOME.NETWORKS:x.ruv.io*relay.ruv.io*agentbbs*mcp*claude code[0:📟MENU]SWARMMINDSAFETYNETWORKTOOLS🔎:find a pageMISSION CONTROL✎goal:what do you want done? e.g. add a dark mode toggle to se…research→create (ADRs, SOP)→build→test→validate→secure→benchmark→learn(1) open Missionsgroups[▾expand all ][▸collapse all ]▾▾⌂⌘ʔ⛭⟳WORKFLOWS2 runs·j/k move·b/l column·d inspect⌂⌘ʔ⛭⟳>>🔀:: Claude Code workflow runs and the ruflo sw…WORKFLOWS0:📟MAIN[SWARM]MINDSAFETYNETWORKTOOLS1:🎯Missions2:🏠Overview3:🐝Swarm[🔀WORKFLOWS]b:👑Hive-Mind4:📌Claimsq:✅Approvals[LINK OK]▸sync…keys on[ Palette ][ x ][✦][ r ][ h ][×]WORKFLOWSreading workflow runs…read when this page opens, then on each refresh while it stays open⌂⌘ʔ⛭⟳sample: build-tune-reviewrunning8 agents·2 running·6 done·0 failed·312.0k tok·6m59s▸1 sample: build-tun…2 sample: build-tun…[ ] switch runPHASESTune·latency, cache and memory passes✔1 Build3/3✔tune-latency52.1k tok2m20s▸❯◒2 Tune2/3✔tune-cache9.8k tok45s◒3 Review1/2◒tune-memory44.7k tok2m24sFigures are derived from the journal and transcripts (read up to 3…le-btr·d inspects the agent·u/i switches run[ prev ][ next ][◂phases ][ agents▸][◂run ][ run▸][ Inspect ][ worktrees ][ next template ][▸deeper ][◂back ][▾next ][▴prev ]Stop, message and redirect: the control tab calls the engine's own…MISSION AUTOPILOTautopilot is offidle·wired·0 running·0 done (0 unverified)·0 failed·rea…[ Stop (9) ][ Check now ]Start is the action row (key 8) and asks first·Stop asks nothing⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳<⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳>> Workflows: the run board, sample run, live⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳8 agents·2 running·6 done·0 failed·312.0k tok·7m01s◒3 Review1/2◒tune-memory44.7k tok2m26s>> Phases: j/k move, b/l pick the column⌂⌘ʔ⛭⟳PHASESReview·security and style reviewers✔1 Build3/3◐review-security71.6k tok1m31s❯◐2 Tune2/3✔review-style21.2k tok35s▸◐3 Review1/2⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳✔1 Build3/3▸✔tune-latency52.1k tok2m20s❯◑2 Tune2/3✔tune-cache9.8k tok45s◑3 Review1/2◑tune-memory44.7k tok2m26sINSPECTtune-memoryINSPECTtune-latencyagenta4sample0004·phase Tune·donemodelSonnet 5.5 (claude-sonnet-5-5)tokens52.1k context tokens of its latest requestelapsed2m20s·started 18:25:30Zworktree/work/demo/.worktrees/tune-latencytool calls4·last Bashresult: p95 latency 212 ms -> 148 ms by batching the two sequentia…transcript/tmp/demo-cfg/cfg.6gzM/projects/-tmp-demo/5a3c…[ Open transcript ]8 agents·2 running·6 done·0 failed·312.0k tok·7m04s❯◒2 Tune2/3✔tune-cache9.8k tok45s◒3 Review1/2◒tune-memory44.7k tok2m29s⌂⌘ʔ⛭⟳❯◐2 Tune2/3✔tune-cache9.8k tok45s◐3 Review1/2◐tune-memory44.7k tok2m29s⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳>> Inspect: model, worktree, tokens, time⌂⌘ʔ⛭⟳❯◑2 Tune2/3▸✔tune-cache9.8k tok45s◑3 Review1/2◑tune-memory44.7k tok2m29sINSPECTtune-cacheagenta5sample0005·phase Tune·donemodelHaiku 4.5 (claude-haiku-4-5)tokens9.8k context tokens of its latest requestelapsed45s·started 18:25:35Zworktreenonetool calls2·last Bashresult: No changes made⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳❯◓2 Tune2/3✔tune-cache9.8k tok45s◓3 Review1/2▸◓tune-memory44.7k tok2m29sagenta6sample0006·phase Tune·runningtokens44.7k context tokens of its latest requestelapsed2m29s·started 18:28:05Zworktree/work/demo/.worktrees/tune-memorytool calls5·last BashStop, message and redirect: the control tab calls the engine's own…/⌂⌘ʔ⛭⟳◒3 Review1/2▸◒tune-memory44.7k tok2m29s⌂⌘ʔ⛭⟳8 agents·2 running·6 done·0 failed·312.0k tok·7m07s◒3 Review1/2▸◒tune-memory44.7k tok2m32selapsed2m32s·started 18:28:05Z⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳◓3 Review1/2▸◓tune-memory44.7k tok2m32s[ prev ][ next ][◂phases ][ agents▸][◂run ][ run▸][ Closetabs[●detail ][○protector ][○mission ][○guide ][○control ]Anatolenot installedpermissionsyour settings allow every classspendhour n/a/n/a·day n/a/n/a·total n/a/n/a (le…journal0 kB of 1500 kB capno envelope approved yet: set one below and press Startno adaptation yet: it needs 6 or more verified steps on a setting…next envelope (a change asks again):classes[●read ][●edit ][●test ][●git-local ][○git-branch ][○network ][○spawn ][○mcp ]hour[−] $2 [+]day[−] $10 [+]total[−] $40 [+]at once[−] 1 [+]duration[−] 7 d [+][○accept running without Anatole ][ load draft file ].claude-flow/console/autopilot/envelope.draft.json holds paths, r…draft is valid·hash ae2ec683d57d·/tmp/demoInside the envelope autopilot does not ask this console. It never…It is not a sandbox (the envelope gates what a task says) and not…AUTOPILOT, PARKED FOR YOUautopilot has not started: nothing is parked>> Control tab: Stop asks first (then we cancel)proof: verified·TaskStop worked on a real background agent (t…[ stop the whole run sample: build-tune-review ]call: TaskStop {"task_id":"wf_sample-btr"}proof: unverified·TaskStop with a workflow run's id was not t…[ message the agent ]call: none (type the guidance first (the field above))proof: queued-only·SendMessage queued a message for a running…[ redirect: stop, then resume with guidance ]proof: unverified·Half one is the stop (see above); half two…A refusal by the engine is shown and nothing else is tried in its…[ stop agent tune-memory ]call: TaskStop {"task_id":"a6sample0006"}tabs[○detail ][○protector ][○mission ][○guide ][●control ]runsample: build-tune-reviewagenttune-memorytool callsbound: each call goes through the engine's per…message:what to tell the agent, or what to change on a redirectmessage: none typed yet▶CONFIRM NEEDEDclick Yes or press yConfirm: stop agent tune-memory?runs: TaskStop {"task_id":"a6sample0006"}Effect: Calls the engine's TaskStop through the session's ownpermission check and dialog. Proven on a background Agent-toolagent; for an agent of a workflow run the engine's answer is theproof. Stopped work is not undone by resuming: a resumed runstarts its unfinished agents again.[ Yes, run it (y) ][ Cancel (n) ]elapsed2m41s·started 18:28:05Zelapsed2m44s·started 18:28:05ZDRILL-DOWN>> Mission autopilot: envelope draft, not started[●network ][○spawn ][○mcp ]draft is valid·hash 6fbfd921accf·/tmp/demo>> Kill switch: Stop (9) is always there; Start asks[ Drill into this run ]e deeper·s search—runs›phases›agen…TRIAGE1 failed·1 empty·4 ok·2 running or waiting✖review-style·Review·error·Error: lint config .eslintrc.jso…○tune-cache·Tune·empty·its whole result is "No changes made"re-run just the failed items (this text is for you to run; theconsole does not):Stop the run first (the Control tab's Stop asks first and callsTaskStop; unverified for a workflow run's own id: Claude Code'sWorkflows panel always works), then:Workflow({ scriptPath: "<the script that started this run>",resumeFromRunId: "wf_sample-btr" })stays cached, its error is its recorded result: review-style(Review). Change that agent() call or its input to run it againClasses come from the run record, the journal's results and thetranscripts; no timeout event is written, so timeout is a textmatch.SEARCHsearch:2+ characters: a run, label, tool call, transcript text or m…Type to search every level at once; Enter jumps to the first hit.>> Parked queue: empty, the autopilot never startedCOSTpricesruflo-cost-tracker book as of 2026-10-03·USD…run$1.00·859.8k billed tokens (all request buck…>> Drill-down: run > phase > agent[ Runs: sample: build-t…]›[ Phases: Tune ]›Agents: tune-memory[▸deeper ][◂back ][▴prev ][▾next ]tune-latencydone·4 calls·2m20stune-cachedone·2 calls·45s▸tune-memoryrunning·5 calls·2m53skeys: e deeper·z back·m next·a prev·q next tab·f follow…[ Runs: sample: build-t…][ Phases: Tune ]…[ Agents: tune-memory ]Agent·Activityview[●Activity ][○Log ][○Files ][○Result ]tune-memory·running·44.7k tok·claude-sonnet-5-5transcript: whole file, 4 KB of the 3.0 MB read cap·12 entries·…▸#1✔Bashnode bench/memory.js --baseli…+0ms900ms#2✔Readsrc/pool.ts+21.8s900ms#3✔Grepnew Array+43.7s900ms#4✔Editsrc/pool.ts+65.5s900ms#5✔Bashnode bench/memory.js+87.3s900ms0 later·call 1 of 5·open it for the full input and output>> Agent tabs: Activity, Log, Files, ResultLogview[○Activity ][●Log ][○Files ][○Result ][●following ][ level: all lines ][ scope: this agent ]18:28:05userSAMPLE RUN—tune-memory: tune step of the d…18:28:16assiStep 1: Bash on the demo project.18:28:16toolBash node bench/memory.js --baseline18:28:38assiStep 2: Read on the demo project.18:28:38toolRead src/pool.ts18:29:00assiStep 3: Grep on the demo project.18:29:00toolGrep new Array18:29:21assiStep 4: Edit on the demo project.18:29:21toolEdit src/pool.ts18:29:43assiStep 5: Bash on the demo project.18:29:43toolBash node bench/memory.js▸18:30:16assiWorking on it.0 above·0 below·line 12 of 12follow pins the newest line and is re-drawn on this page's refresh…Filesview[○Activity ][○Log ][●Files ][○Result ]worktree /work/demo/.worktre…[ Read diff stat ]diff stat not read: the worktree is not inside this project's fold…src/pool.tsedit×1 read×1Resultview[○Activity ][○Log ][○Files ][●Result ]The agent's return is not in memory (the run folder is outside Cla…[ Read the journal (up to 1.0 MB) ][ Runs: sample: build-t…][ Phases: Tune ][ Agents: tune-memory ][[▸deeper ][◂back ][▴prev ][▾next ]Agentool call Bash·#3 of 12·18:28:16·ok·900ms·input·50 characters·lines 1–3 of 3Acti{ity"command": "node bench/memory.js --baseline"}output·61 characters·lines 1–3 of 3rss 171 MBDEMO_API_‹masked›heap: pool=4096[▴up ][▾down ]6 lines in all>> A call and its output: secrets are masked▸tune-memoryrunning·5 calls·3m02sTranscripts: up to 3.0 MB each (a larger one is read from itsend:≥); at most 24.0 MB parsed per refresh; 8 of 16 read.≥▸tune-memoryrunning·5 calls·3m05smarks a floor,≈a family price.Build$0.39306.1k tok>> Triage: failed, empty, and the re-run text▸Tune$0.28332.4k tokReview$0.32221.4k toktune-latencySonnet 5.5$0.14156.3k toktune-cacheHaiku 4.5$0.00919.6k toktune-memorySonnet 5.5$0.14156.4k tokguardsoff: set wfBudgetRunUsd, wfBudgetDayUsd or wfA…MISSION LINKS0 agent-to-task links·16 agents unlinked·0 mission tasks knownNo mission yet in the ledger or the observation: make one in Missi…agents unlinked: its label has no [mission:<id> task:<id>] tag (a…WORKTREES (READ-ONLY LIST; ONE CONFIRM-GATED REMOVAL)3 worktrees·read 46s ago·origin/main as last fetched 3d ago (…….worktrees/…↑0↓0·clean·1m…kept: outside .claude/worktrees an…….worktrees/…↑1↓0·clean·1m…kept: outside .claude/worktrees an…>> Search: one term, every level at once3 worktrees·read 47s ago·origin/main as last fetched 3d ago (…search:2+ characters: a run, label, tool call, transcript text…⏎gosearch:pool⏎go6 matches·Enter in the field goes to the firstAgents (1)[▸tune-memorysample: build-tune-revie…]Tool calls (2)[▸Read src/pool.tstune-memory·call #2·…][▸Edit src/pool.tstune-memory·call #4·…]Transcript text (3)[▸rss 171 MB DEMO_API_‹masked›…tune-memory·output of…][▸class Pool { /* 96 lines */ }tune-memory·output of…][▸src/pool.ts:41: new Array(40…tune-memory·output of…]looked through 8 agents' transcripts (3,435 characters of 8,000,00…8 agents' transcripts are not in memory, so they were not searched[ Read this run's transcripts (up to 24 MB in all) ]3 worktrees·read 50s ago·origin/main as last fetched 3d ago (……tmp/demo…↑0↓0·clean·1m·not att…kept: the main wo…live processes: not checked (the read button, w, checks them; the…3 worktrees·read 51s ago·origin/main as last fetched 3d ago (…cleanup: live processes not checked: press the read button firstremoved: the directory only, never the branch, never with --force…3 worktrees·read 52s ago·origin/main as last fetched 3d ago (…[ read worktrees + live processes ]>> Cost: per run, phase, agent (a price list)3 worktrees·read 53s ago·origin/main as last fetched 3d ago (…3 worktrees·read 54s ago·origin/main as last fetched 3d ago (…SWARM NESTINGPick the ruflo swarm run ([ ] to switch) to see how its swarm, hiv…REPLAY[▸open ]14 moments in 6m10s: scrub back and forth, jump by phase…WORKFLOW TEMPLATES (DRY RUN, THEN A CONFIRMED LAUNCH)3 worktrees·read 55s ago·origin/main as last fetched 3d ago (…te…[●Review ][○Migrate ][○Research ][○Build, tune, review ]Independent reviewers read the same change through one lens each, averifier challenges what they found, one agent writes it up.what to review: the current uncommitted diffreviewers[−] 4 [+](2 to 8)>> Mission links: no mission yet, so none linked3 worktrees·read 56s ago·origin/main as last fetched 3d ago (…3 worktrees·read 57s ago·origin/main as last fetched 3d ago (…lens…[●correctness ][○security ][○performance ][○all ]dry run: 7 agents in 4 phases (Gather 1·Review 4·Verify 1·R…3 worktrees·read 58s ago·origin/main as last fetched 3d ago (…counts only: tokens, time and cost are not estimated, nothing has…launch (the action row) sends this as a visible prompt after you…3 worktrees·read 59s ago·origin/main as last fetched 3d ago (…>> Worktrees: read-only list (sample worktrees)3 worktrees·read 1m ago·origin/main as last fetched 3d ago (n…[ reading…]3 worktrees·read 1s ago·origin/main as last fetched 3d ago (n……tmp/demo…↑0↓0·1 changed·1m·3 live·…kept: the main…live processes: 690 working directories seen, 1s ago, your own pr…cleanup: nothing is both merged and cleanRUN A MISSION AS A WORKFLOWlive processes: 690 working directories seen, 2s ago, your own pr…No active mission: create one in Mission Control, and its SPARC pl…COMPARE RUNS[▸open ]1 other run of "sample: build-tune-review": agents, toke…PROJECT ANATOLE, PER AGENT (REPORTED BY THE MOD)Project Anatole is not installed or has written nothing here: cla…>> Swarm nesting: needs a ruflo swarm runEXPORT RUN SUMMARY[▸open ].claude-flow/console/exports/sample-build-tune-review-wf…>> Replay: step through the finished sample run3 worktrees·read 6s ago·origin/main as last fetched 3d ago (n…live processes: 690 working directories seen, 6s ago, your own pr…3 worktrees·read 7s ago·origin/main as last fetched 3d ago (n…live processes: 690 working directories seen, 7s ago, your own pr…3 worktrees·read 8s ago·origin/main as last fetched 3d ago (n…▸tune-memoryrunning·5 calls·3m32sidle·not started·0 running·0 done (0 unverified)·0 failed…◒3 Review1/2▸◒tune-memory44.7k tok3m35s8 agents·2 running·6 done·0 failed·312.0k tok·8m10s◑3 Review1/2▸◑tune-memory44.7k tok3m35s⌂⌘ʔ⛭⟳[ Runs: sample: build-t…]›[ Phases: Build ]›Agents: api-core▸api-coredone·9 calls·2m30scli-flagsdone·6 calls·1m30sdocs-passdone·4 calls·1m00s1 empty·7 okrunn/a·n/a billed tokens (all request buckets)…▸Buildn/an/aTunen/an/aReviewn/an/aapi-coreOpus 5.5n/an/acli-flagsSonnet 5.5n/an/adocs-passHaiku 4.5n/an/a3 worktrees·read 18s ago·origin/main as last fetched 3d ago (…live processes: 690 working directories seen, 18s ago, your own p…3 worktrees·read 19s ago·origin/main as last fetched 3d ago (…live processes: 690 working directories seen, 19s ago, your own p…[▸open ]16 moments in 9m20s: scrub back and forth, jump by phase…live processes: 690 working directories seen, 20s ago, your own p…[▾close ]16 moments in 9m20s: scrub back and forth, jump by phas…step 0/16·+0s of +9m20s·16x·⏸paused▸before the first agent started[⏮][◂◂phase ][◂step ][▶play ][ step▸][ phase▸▸][⏭]speed[−]16x[ + ]of 1·4·16·64·256○Build0/3○api-corequeuedn/an/a○cli-flagsqueuedn/an/a○docs-passqueuedn/an/a○Tune0/3○tune-latencyqueuedn/an/a○tune-cachequeuedn/an/a○tune-memoryqueuedn/an/a○Review0/2○review-securityqueuedn/an/a○review-stylequeuedn/an/aTokens show n/a until an agent ends: only its final figure is rec…This replay cannot rewind a run; stop or message one from its con…step 1/16·+0s of +9m20s·16x·⏸paused▸api-core started (Build)◐Build0/3·1 running◐api-corerunningn/a0sCONVERSATION[▸open ]talk to Claude, its agents, the hive, federation, Codex…>> Templates: dry run counts, nothing launchedte…[○Review ][○Migrate ][○Research ][●Build, tune, review ]Builders each write one part in their own worktree, tuners measureand improve it, two reviewers read the result.what to build:the feature and where it goes (required)builders[−] 3 [+](1 to 6)tuners[−] 1 [+](1 to 4)dry run: 6 agents in 3 phases (Build 3·Tune 1·Review 2), at m…needs: what to buildSAVED VIEWS AND PINS[▸open ]0 pinned·the cursor is remembered across restartsread 1s ago·re-read on each refresh while this page is open, nev…>> Run a mission as a workflow: no mission, so empty>> Compare runs: this sample run vs yesterday'sread 3s ago·re-read on each refresh while this page is open, nev…[▾close ]1 other run of "sample: build-tune-review": agents, tok…A prevvsB le-btr[ changed only ]stateA completedB runningagentsA 8 (8 ok, 0 failed) B 8 (6 ok, 0 failed)tokensA 374.5kB 312.0k-17%timeA 9m25sB 8m31s-10%1 same·7 changed·0 only in B (+)·0 only in A (-)~ = toke…~ api-coreA✔70.2k 2m30sB✔61.4k 1m50s…= cli-flagsA✔41.8k 1m30sB✔38.9k 1m15s…~ docs-passA✔15.1k 1m00sB✔12.3k 55s…~ tune-latencyA✔63.9k 2m50sB✔52.1k 2m20s…~ tune-cacheA✔11.4k 50sB✔9.8k 45s…~ tune-memoryA✔58.3k 3m50sB◐44.7k 3m56s…~ review-securityA✔80.8k 2m20sB◐71.6k 3m01s…~ review-styleA✔33.0k 1m20sB✔21.2k 35s…run B is derived from journal and transcripts, not a finished run…result text differs between the two (it always can; only the agen…timeA 9m25sB 8m34s-9%~ tune-memoryA✔58.3k 3m50sB◐44.7k 3m59s…~ review-securityA✔80.8k 2m20sB◐71.6k 3m04s…read 2s ago·re-read on each refresh while this page is open, nev…>> Export run summary: default path (not exported)read 0s ago·re-read on each refresh while this page is open, nev…[▾close ].claude-flow/console/exports/sample-build-tune-review-w…Writes a markdown summary: phases, agents, tokens, time, results…Allowed: under /tmp/demo·name ends in .md·no ..·never overw…path:.claude-flow/console/exports/sample-build-tune-review-wf_samp…[ Export to the default path ]>> Conversation: targets and composer, nothing sent[▾close ]talk to Claude, its agents, the hive, federation, Codex…[●@claude ]Claude (this session)·prompt·leaves: machine·se…sends: nothing beyond what a turn of this session sends to th…cost: a visible turn of this session, billed as any turn isreply: its answer is the next turn in the transcript, not cap…[○@tune-memory ]agent tune-memory·send-message·leaves: machi…[○@review-security ]agent review-security·send-message·leave…[○@task ]ruflo task note·task·leaves: machine·free[○@peer-alpha1 ]peer alpha1·peer·leaves: tailnet·peer-sess…[○@peer-bravo-1 ]peer bravo-1·peer·leaves: tailnet·peer-se…[○@peer-charlie-node ]peer charlie-node·peer·leaves: tailnet…[○@peer-delta-hub-0 ]peer delta-hub-0·peer·leaves: tailnet·…[○@peer-delta-hub-1 ]peer delta-hub-1·peer·leaves: tailnet·…[○@peer-echo-worker-node-1 ]peer echo-worker-node-1·peer·lea…[○@peer-echo-worker-node-2 ]peer echo-worker-node-2·peer·lea…[○@peer-echo-worker-node-3 ]peer echo-worker-node-3·peer·lea…[○@peer-foxtrot-v0 ]peer foxtrot-v0·peer·leaves: tailnet·p…[○@peer-hotel-cluster-01 ]peer hotel-cluster-01 ·peer·leave…[○@peer-hotel-cluster-02 ]peer hotel-cluster-02 ·peer·leave…[○@peer-hotel-cluster-03 ]peer hotel-cluster-03 ·peer·leave…[○@peer-relay-01 ]peer relay-01·peer·leaves: tailnet·peer-…[○@peer-leaf-1 ]peer leaf-1·peer·leaves: tailnet·peer-sess…[○@peer-leaf-2 ]peer leaf-2·peer·leaves: tailnet·peer-sess…[○@peer-leaf-3 ]peer leaf-3·peer·leaves: tailnet·peer-sess…[○@peer-leaf-4 ]peer leaf-4·peer·leaves: tailnet·peer-sess…[○@peer-leaf-5 ]peer leaf-5·peer·leaves: tailnet·peer-sess…[○@peer-india-edge-node-1 ]peer india-edge-node-1·peer·leave…[○@codex ]Codex (codex exec, read-only)·codex·leaves: intern…[○@openrouter ]openrouter (openrouter/auto)·http·leaves: int…message:@claude @openrouter what do you think? (@all = agents, hive…>> Saved views and pins: the cursor is remembered[▾close ]0 pinned·the cursor is remembered across restartsFile .claude-flow/console/wf-views.json (≤64 kB, versioned)·no…Remembers the cursor (run, phase, agent by name), filters and sea…search: none·filters: none[ Pin this run ][ Forget saved views ]No pinned runs (up to 20).….worktrees/…↑1↓0·clean·2m…kept: outside .claude/worktrees an……tmp/demo…↑0↓0·1 changed·2m·3 live·…kept: the main…live processes: 690 working directories seen, 57s ago, your own p…….worktrees/…↑0↓0·clean·2m…kept: outside .claude/worktrees an…live processes: 690 working directories seen, 58s ago, your own p…live processes: 690 working directories seen, 59s ago, your own p…Figures are from the run record Claude Code wrote when it finished.prev·d inspects the agent·u/i switches runsample: build-tune-reviewcompleted8 agents·0 running·8 done·0 failed·374.5k tok·9m25s1 sample: build-tun…▸2 sample: build-tun…[ ] switch runPHASESBuild·client, CLI flags, docs in parallel▸✔1 Build3/3✔api-core70.2k tok2m30s✔2 Tune3/3✔cli-flags41.8k tok1m30s✔3 Review2/2✔docs-pass15.1k tok1m00s⌂⌘ʔ⛭⟳⌂⌘ʔ⛭⟳[ Runs: sample: build-t…]›[ Phases: Tune ]›Agents: tune-latency▸tune-latencydone·4 calls·2m20stune-memoryrunning·5 calls·4m26stune-memoryrunning·5 calls·4m29s>> sample run, not real data>> ruflo console: the Workflows board (sample run)RUFLO CONSOLE: THE WORKFLOWS BOARD (SAMPLE RUN)R U F L OC O N S O L Ev0.34.1the WORKFLOWS boardphases◂▸agents·drill-down·searchtriage·cost·control·autopilotevery figure comes from the run files:nothing is estimatedSAMPLE RUN: authored for this demo/plugin install ruflo-console@ruflothen/ruflo workflows