Add SendMessage continuation protocol to plan-review subagent dispatches#3371
Merged
Trecek merged 2 commits intoMay 31, 2026
Merged
Conversation
…ispatches - Bump plan-registry-tracer maxTurns from 40 to 80 (reduces continuation frequency) - Add SendMessage continuation protocol paragraph to Steps 5/6/7 in rectify/SKILL.md - Add SendMessage continuation protocol paragraph to Steps 6/7/8 in make-plan/SKILL.md - Add T1 (maxTurns >= 80) to tests/server/test_tools_agents.py - Add T2/T3 (SendMessage protocol presence) to tests/skills/test_adversarial_review_contracts.py - Update tests/skills/CLAUDE.md description for test_adversarial_review_contracts.py Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
…continuation blocks All three rectify continuation protocol blocks (Foundation Audit, Interface Mapping, Registry Trace steps) carried "Continue plan review subagent to finalize findings" — make-plan-specific wording copied verbatim in the same commit that introduced the protocol. Changed to "Continue rectify subagent to finalize findings" to correctly name the executing skill context. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Trecek
deleted the
rectify-make-plan-sendmessage-continuation-protocol-missing/3350
branch
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
When a plan-review subagent (foundation auditor, interface mapper, or registry tracer) exhausts its
maxTurnslimit, the Claude Code harness returns a continuation hint with anagentId. Neither rectify nor make-plan SKILL.md files provide guidance for this case, causing orchestrators to improvise — and get theSendMessageschema wrong (omitting the requiredsummaryfield). This plan bumps the registry tracer'smaxTurnsfrom 40 to 80 (reducing continuation frequency) and adds an explicit SendMessage continuation protocol to every adversarial subagent dispatch step in both skills.Closes #3350
Implementation Plan
Plan file:
/home/talon/projects/autoskillit-runs/impl-20260530-183836-227844/.autoskillit/temp/make-plan/rectify-make-plan-sendmessage-continuation-protocol-missing_plan_2026-05-30_184600.mdChanged Files
src/autoskillit/agents/plan-registry-tracer.mdsrc/autoskillit/skills_extended/make-plan/SKILL.mdsrc/autoskillit/skills_extended/rectify/SKILL.mdtests/server/test_tools_agents.pytests/skills/CLAUDE.mdtests/skills/test_adversarial_review_contracts.py🤖 Generated with Claude Code via AutoSkillit
Token Usage Summary
* Step used a non-Anthropic provider; caching behavior may differ.
Token Efficiency
Model Usage Breakdown