Recognize agent-orchestrated sessions in transcript policy#136
Merged
Conversation
AGENTS.md's transcript policy assumed the human-with-Claude workflow where every session has real `### User` messages. Conductor's parallel- launch model breaks that assumption: the implementation session has no direct user, only an agent-orchestrated launching prompt. Add a new "Agent-orchestrated sessions" subsection to AGENTS.md and a matching note to the Feature PR Documentation Bundle AI check. The new convention: use `### Parent agent (orchestrator)` headings instead of `### User`, reproduce the parent-agent prompt verbatim, and declare the session as agent-orchestrated at the top of `## Detailed conversation`. Same verbatim rule applies — summarized parent prompts are still rejected. This eliminates a recurring false-negative on the docs check that was raised on PRs #129, #130, #131, #133. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
rafacm
added a commit
that referenced
this pull request
May 4, 2026
Updates the planning and implementation session transcripts to comply with the new "Agent-orchestrated sessions" subsection in AGENTS.md (merged in #136 / 93056e3). - Planning transcript: replace summarized user sections with verbatim text from the parent conversation that authorized this work. - Implementation transcript: declare the session as agent-orchestrated at the top of `## Detailed conversation`, replace the `### User` heading with `### Parent agent (orchestrator)`, and reproduce the parent agent's launching prompt verbatim. - Session ID values audited; placeholder `unavailable` used where the real UUID is unrecoverable, replacing duplicated/fabricated values. This is a verification PR for the new convention — the Feature PR Documentation Bundle AI check should flip from FAIL to PASS after this commit. The Comment Discipline check (long explanatory blocks in episodes/apps.py) remains failing and is intentionally deferred to a separate follow-up. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
rafacm
added a commit
that referenced
this pull request
May 4, 2026
Updates the planning and implementation session transcripts to comply with the new "Agent-orchestrated sessions" subsection in AGENTS.md (merged in #136 / 93056e3). - Planning transcript: replace summarized user sections with verbatim text from the parent conversation that authorized this work. - Implementation transcript: declare the session as agent-orchestrated at the top of `## Detailed conversation`, replace the `### User` heading with `### Parent agent (orchestrator)`, and reproduce the parent agent's launching prompt verbatim. - Session ID values audited; placeholder `unavailable` used where the real UUID is unrecoverable, replacing duplicated/fabricated values. This is a verification PR for the new convention — the Feature PR Documentation Bundle AI check should flip from FAIL to PASS after this commit. The Comment Discipline check (long explanatory blocks in episodes/apps.py) remains failing and is intentionally deferred to a separate follow-up. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
rafacm
added a commit
that referenced
this pull request
May 4, 2026
Updates the planning and implementation session transcripts to comply with the new "Agent-orchestrated sessions" subsection in AGENTS.md (merged in #136 / 93056e3). - Planning transcript: replace summarized user sections with verbatim text from the parent conversation that authorized this work. - Implementation transcript: declare the session as agent-orchestrated at the top of `## Detailed conversation`, replace the `### User` heading with `### Parent agent (orchestrator)`, and reproduce the parent agent's launching prompt verbatim. - Session ID values audited; placeholder `unavailable` used where the real UUID is unrecoverable, replacing duplicated/fabricated values. This is a verification PR for the new convention — the Feature PR Documentation Bundle AI check should flip from FAIL to PASS after this commit. The Comment Discipline check (long explanatory blocks in episodes/apps.py) remains failing and is intentionally deferred to a separate follow-up. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
8 tasks
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
AGENTS.md's transcript policy assumed the human-with-Claude workflow where every session has real
### Usermessages. Conductor's parallel-launch model breaks that assumption: the implementation session has no direct user, only an agent-orchestrated launching prompt.This PR adds a new "Agent-orchestrated sessions" subsection to AGENTS.md and a matching note to the Feature PR Documentation Bundle AI check.
The new convention:
### Parent agent (orchestrator)headings instead of### User## Detailed conversationThis eliminates a recurring false-negative on the docs check that was raised on PRs #129, #130, #131, and #133 — where the implementation agents had no real user to quote and the existing policy left them no honest path forward.
Classification
Doc-only / policy clarification PR. Per the AI Checks rule's exclusions, this does not need the full feature documentation bundle (plan + feature doc + both transcripts). CHANGELOG entry only.
Test plan
🤖 Generated with Claude Code