Repository navigation
feat(j5): agents choose between subagents and Crew members from context - #758
bryantderosier merged 16 commits into
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
3bd2d5c to
1d4ff9b
Compare
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars.
|
cd845be to
44d7306
Compare
44d7306 to
e56c45c
Compare
e56c45c to
4225604
Compare
Panel review: #758Should fix: the "only" rules contradict the final-result and step-report rules. Three places say the Captain hears from a member only for a review pass, a blocker or a disagreement:
The seat brief ( Nit: race in Docs nits
CI: this head's one failure is Review panel for the #750–#762 stack, judged as the code that lands at |
4225604 to
e5c8e7b
Compare
…746) Rewrites crews.md AC1 and AC9, adds the delivery outcome as AC24 and a per-provider steering table (live-test results pending), updates the quoted tool descriptions in agent-tools.md, FORK.md cases 8 and 26, the register's D1, and fixes two stale lines (persona parameter name, delegate_task in the glossary). Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
#746) The three places that said a Captain hears from a member only for a review pass, a blocker or a disagreement contradicted the seat brief and the playbook hand-off, which both have members report a final result or a step. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
…Claude steering (#746) The live-test results belong to the PR. The user doc still said every provider but Codex GPT 6 queues, which Claude's steering contradicts. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
d3926de to
380adfa
Compare
|
Review panel: Claude Opus 5.5 (lead) and GPT-6.1 Sol. We reviewed the whole stack at Low should-fix: the A steer that was accepted counts as delivered, and there's no later receipt that the model processed it. Your own live test shows a steered message lost to a Stop on Claude, with no alarm (FORK case 26). An agent told Suggestion: say that |
|
Jackson's decision, relayed by the review lead (Claude Opus 5.5): "crew", "peer agent" and "subagent" stay trigger words. In Jackson's words: "If I ask for a feature, I'd expect to get it." The new standing instructions say "the words in the user's message do not decide it, and 'crew' or 'subagent' in that message does not by itself pick one", and Please change both:
crews.md AC1, D1 and the test pins follow. |
|
Jackson's decision, relayed by the review lead (Claude Opus 5.5): please remove The value is predicted when the message is sent. Your own live test had a message reported as Please:
This also settles the earlier panel finding on this PR about the description promising an alert. |
…-Crew layer The standing instructions are ported onto j5/main's section layout. Per the review, an explicit request for a Crew, a Peer Agent or a subagent gets that mechanism, and the choice from the work applies only when the person asked for none. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
…op predicting delivery Jackson's review decisions on this layer: an explicit request for a Crew, a Peer Agent or a subagent gets that mechanism and the choice from the work applies only when the person asked for none; and j5_send_message no longer returns a predicted deliveryOutcome, keeping the measured deliveryNotice, the offline-recipient delivery field and the late-failure alarm. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
…-Crew layer Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
|
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
…ing exported Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Summary
Agents have too many ways to get help, and the guidance reads like a menu, so they reach for a Crew because I said "crew" or spawn a standalone Peer Agent when a subagent would do. Inside a Crew, review rounds also go through the Captain, with acknowledgments and progress chatter in between, and a member that sends something important can't tell whether it landed. This rewrites the guidance around the work, makes members run their own review loops, and makes
j5_send_messagesay what happened to the message. Closes #746.This is stacked on #770, so the base is
j5/crew-member-reasoning-levels. It sits above #744 and #770 on purpose: the tool descriptions I rewrite here already carry their model and reasoning-level eligibility sentences. Three more PRs (#760, #761, #762) are stacked on it.orchestrationInstructions.ts) now offer three choices, one sentence each: work directly, subagents for independent assignments, or a Crew for work where workers have to respond to each other. An explicit ask for a Crew, a Peer Agent or a subagent still gets that mechanism, and the choice from the work only applies when nobody asked (Jackson's call in review). An agent can reassess mid-work without redoing finished work, and the test pins and FORK.md case 8 changed with it.delegate_taskacross providers), or members on a Crew (j5_propose_crewto start one,j5_request_crew_memberto grow one). Each tool description says which path it serves.j5_spawn_agentis now described only for a person's explicit request for a standalone agent, or when an outside agent or a skill names it. Nothing a caller depends on was removed.<crew_collaboration>tells members to send findings and fixes straight to the teammate who owns the work, repeat until none are unresolved, and message the Captain only for a review pass (naming the commit reviewed), a blocker, or a disagreement they couldn't settle. No acknowledgments or progress updates. A loop that hasn't produced a clean pass after three rounds goes to the Captain. The Captain settles technical disagreements within the agreed scope and asks me only when it needs product intent or my judgment. The Captain also hears a member's final result or step report, so every "only" list in the instructions, thej5_propose_crewdescription, the launch report and the seat brief says so.agent-personas/index.mdsaidj5_spawn_agenttookagent(it'spersona), andglossary.mdsaiddelegate_taskwasn't on the J5 agent surface.Callers of the tools I touched. Nothing a caller depends on was removed. These are the ones I found in the repo; personas, playbooks and scheduled tasks that people wrote themselves are user data and can't be inventoried from here:
orchestrationInstructions.ts,agentInvocationInstructions.ts(delegate_taskfor@persona:),playbookInstructions.ts(j5_propose_crewfor playbook Crews). Pi's extension takes the same text through upstream's constant.mcp/tools.ts(spawn, propose, request member, list personas),agentDelegation.ts(delegate_task),crewGateNotice.ts,spawnIds.ts(<crew_collaboration>), plus refusal hints inoutsideAgent.tsandhandlers.ts, and mentions in the worktree and project toolkits.j5ToolPreapproval.ts,agentPersonaRuntime.ts.j5-live-testskill still tells the builder to spawn a tester withj5_spawn_agent, which the new text allows because a skill names it.Docs: crews.md (AC1 and AC9 rewritten, and a messages-and-steering table), agent-tools.md including the quoted descriptions,
docs/j5/product/upstream.mdD1, and a sentence in the personas user guide.Evidence
At
dfe913afdc, on top of #770 and j5/main. #744's reasoning levels moved into #770 below it, and this layer's own tests didn't change:cd apps/server && vp test run src/j5/a2a src/j5/agents src/mcpPeerPollRoundTrip.test.tsstill covers the offline-recipient case, and theEnvelopeFormattertest checks thej5_send_messagedescription against agent-tools.md byte for byte.orchestrationInstructions.test.tsandplaybookInstructions.test.ts:cd packages/provider-core && vp test runspawnIds.test.ts,tools.test.tsandcrewGateNotice.test.tspin the new<crew_collaboration>, tool-description and Captain wording.Not verified: Cursor's steering (it isn't set up on the test host) and Codex models other than GPT 6. The Live testing section below has what ran. No repo-wide checks.
Merge Danger
Door: two-way. No migration, no stored data, and a revert restores the old text and the old
deliveryfield.Blast Radius: every agent's standing instructions. Agents will choose between subagents and Crews differently. The
j5_send_messageresult is unchanged.PLAYBOOK_CREW_INSTRUCTIONSso "crew" in a message isn't the trigger for building a playbook Crew.After merge: nothing.
Live testing
Ran on this stack at
1d4ff9bdf6. Later commits only add the results tocrews.md, and the commits below it were reworked (the settings row) and rebased onto a newer j5/main, which this layer doesn't touch. Isolated test home fromj5-live-test, Claude Haiku 5.5 for the Captain and writer, GPT 6 Luna for the reviewer. I checked each result in the member's own thread and the ledger rather than the Captain's account.j5_send_messagereportedstartingand the writer started a new turn.scripts/hold.sh, a second message reportedsteeredand showed up inside that running turn in the writer's thread.hold.shwas gone. The steered message was already marked delivered, but the writer never acted on it, so it was lost with the turn. That's the known race in FORK case 26, and Claude shows it.starting, and the writer did the work in a new turn.steered, and the reviewer acted on it mid-run: it sent its review before I stopped it. Stop interrupted cleanly and every delivery ended up delivered.Not exercised live: Cursor (not set up here), Codex models other than GPT 6, a message after the run ended on Codex, and a steered message to a member on an online peer server. The per-provider results are in the list above, not in
crews.md, which only states what each provider does.Upstream impact
FORK.mdcases 8 and 26 are updated. Upstream's constant already points at J5's instruction text, so no upstream source file changes.docs/j5/product/upstream.mdD1 is rewritten for the new wording.Checklist
FORK.md(case text and file-table row) in this PRdocs/j5/product/upstream.md— D1 rewrittenAGENTS.md) — Claude and Codex steering ran live; Cursor's is not live-tested, as the Live testing section saysdocs/j5/product/and user docs rewritten where this changes themClaude Sonnet 5.5 through the Claude Code harness in J5 Code.
🤖 Generated with Claude Code