fix: credit operator-decided items in report tally + footer (#5) - #7
Merged
Conversation
`impasse_report.py show` reported "Nothing needed you — the models settled all N between themselves" even for runs where the operator personally decided a finding, because the tally and footer keyed off state == "deadlocked" rather than the presence of an escalation object. That violates SKILL.md's rule that operator rulings count as escalations regardless of channel. render() now credits operator-decided items (state "resolved" carrying an "escalation" object) separately: - tally: a "🧑⚖️ N decided by you" segment (the "escalated to you" deadlock count is preserved unchanged); - footer: an operator-decided-only run says the models settled the rest and you decided N (not "nothing needed you"); the mixed case (a pending deadlock AND an operator-decided item) credits both instead of implying the operator's ruling was model-settled; the genuinely-autonomous case is unchanged. The "not in scope" behaviors from the issue (_open_escalations, running stats) are untouched. Closes #5. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01X8XaW2bqgYnp7oRVgfHXid
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #5.
impasse_report.py showprinted "Nothing needed you — the models settled all N between themselves" even for runs where the operator personally decided a finding, because the tally and footer keyed offstate == "deadlocked"rather than the presence of anescalationobject — exactly what SKILL.md's "operator rulings count as escalations regardless of channel" rule warns against.Fix (
render()only)Credit operator-decided items (state
resolvedcarrying anescalationobject) separately:🧑⚖️ N decided by yousegment; the⚖️ … escalated to youdeadlock count is preserved unchanged.The issue's "not in scope" behaviors (
_open_escalations, running stats) are untouched.Tests
report: operator-decided (resolved + escalation) item is credited, not counted as autonomousreport: genuinely-autonomous run still says 'Nothing needed you'(regression guard)report: a deadlock still takes footer precedence over an operator-decided item(asserts the mixed footer credits "you decided N")The mixed-case footer fix was surfaced by a cross-provider Impasse review of the branch. Three gates green.
🤖 Generated with Claude Code