Conversation
The `<!-- AUTO-GENERATED ... -->` HTML comment before frontmatter prevented tools like agnix from parsing YAML frontmatter on line 1. Removed the header entirely - the adapters/ directory is self-explanatory. Drops agnix errors from 83 to 8.
… plugins Debate skill hardened based on its own first debate's findings: - Universal evidence standard for both proposer AND challenger - Proposer prompts now require cited evidence (was challenger-only) - Challenger follow-up reordered: anti-convergence guard first - Minimum-disagreement requirement per round added - Context summarization criteria specified (500-800 tokens) - Rigor indicator and Debate Quality rating in synthesis output Consult skill fixes from enhance analysis: - Gemini section: added missing Session ID extraction line - Codex: removed invalid -a suggest flag (codex exec doesn't support it) - Codex: added -c model_reasoning_effort to safe command patterns - Gemini models: replaced all -preview suffixes with stable names Also updates README, test strategy doc, and all adapters.
…, relaxed disagreement Three fixes from the claude-vs-codex debate findings: 1. --context passthrough from /debate command through orchestrator to consult skill, so both tools see the same codebase context 2. Canonical redaction: orchestrator references consult agent's full 14-pattern table instead of maintaining a drifted 5-pattern subset 3. Relaxed forced disagreement: challenger can now certify a concern as genuinely resolved with evidence, not forced to always disagree
…, relaxed disagreement
Fixes from debate findings + enhance/review analysis:
Debate plugin:
- --model=auto bug: omit --model instead of passing "auto" literally
- --context=file: add follow-up path prompt with validation
- --context added to interactive picker
- Template {round-1} -> {previous_round} for clarity
- Relaxed disagreement: can certify concerns resolved with evidence
Consult plugin:
- Codex model: gpt-5-codex-mini/gpt-5.1-codex-mini -> gpt-5.3-codex-spark
- Redaction table: skill is now canonical source (14 patterns), agent references it
- --continue + --count mutual exclusion in agent
- OpenCode model: claude-opus-4-5 -> claude-opus-4-6
Infrastructure:
- Adapter transform: preserve .claude/ in platform doc enumerations
- New: __tests__/debate-command.test.js (76 tests)
- Updated consult tests for new redaction ownership
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
Summary of ChangesHello @avifenesh, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request significantly refines the Highlights
Changelog
Activity
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request introduces a number of valuable fixes and enhancements, particularly for the debate and consult plugins. The move to a canonical source for redaction patterns is a great improvement for maintainability. The new debate logic, which enforces evidence-based arguments and encourages disagreement, is well-designed and should lead to higher quality outputs. The addition of a comprehensive test suite for the debate plugin is also a significant contribution.
My main concern is the significant code duplication of the adapter transformation logic. The changes in lib/adapter-transforms.js have been copied to 11 other plugin-specific adapter-transforms.js files. This creates a maintenance burden and should be refactored to use a single, shared module. I've left a specific comment on this issue.
Contributor
There was a problem hiding this comment.
Pull request overview
This PR applies “debate findings” across the debate/consult markdown plugins and the adapter-generation pipeline, updating context passthrough semantics, model defaults, canonical redaction ownership, and adapter transforms; it also adds extensive markdown-structure tests for the debate plugin.
Changes:
- Update adapter generation to remove the auto-generated header so YAML frontmatter starts on line 1.
- Update OpenCode adapter transforms to preserve
.claude/in platform enumeration docs while still rewriting.claude/paths elsewhere. - Update debate/consult plugins for context passthrough, model selection changes, and canonical redaction documentation; add/adjust Jest tests accordingly.
Reviewed changes
Copilot reviewed 122 out of 122 changed files in this pull request and generated 4 comments.
Show a summary per file
| File | Description |
|---|---|
| scripts/gen-adapters.js | Stop prepending auto-generated header to adapter outputs. |
| lib/adapter-transforms.js | Preserve .claude/ in “Claude Code:” enumerations during OpenCode transforms. |
| plugins/audit-project/lib/adapter-transforms.js | Sync OpenCode .claude/-preservation behavior into plugin-local lib copy. |
| plugins/debate/agents/debate-orchestrator.md | Add context passthrough behavior; point redaction to consult skill as canonical source. |
| plugins/debate/commands/debate.md | Add --context flag + interactive picker and file-path follow-up flow. |
| plugins/debate/lib/adapter-transforms.js | Sync OpenCode .claude/-preservation behavior into plugin-local lib copy. |
| plugins/debate/skills/debate/SKILL.md | Strengthen evidence rules, rename {round-1} to {previous_round}, update synthesis constraints. |
| plugins/consult/agents/consult-agent.md | Add --continue + count>1 mutual exclusion; reference skill as canonical redaction source. |
| plugins/consult/commands/consult.md | Update Codex safe-mode default guidance; update Codex model label. |
| plugins/consult/skills/consult/SKILL.md | Update Gemini/Codex model mappings; move full redaction table into skill (canonical). |
| plugins/deslop/lib/adapter-transforms.js | Sync OpenCode .claude/-preservation behavior into plugin-local lib copy. |
| plugins/drift-detect/lib/adapter-transforms.js | Sync OpenCode .claude/-preservation behavior into plugin-local lib copy. |
| plugins/enhance/lib/adapter-transforms.js | Sync OpenCode .claude/-preservation behavior into plugin-local lib copy. |
| plugins/learn/lib/adapter-transforms.js | Sync OpenCode .claude/-preservation behavior into plugin-local lib copy. |
| plugins/next-task/lib/adapter-transforms.js | Sync OpenCode .claude/-preservation behavior into plugin-local lib copy. |
| plugins/perf/lib/adapter-transforms.js | Sync OpenCode .claude/-preservation behavior into plugin-local lib copy. |
| plugins/repo-map/lib/adapter-transforms.js | Sync OpenCode .claude/-preservation behavior into plugin-local lib copy. |
| plugins/ship/lib/adapter-transforms.js | Sync OpenCode .claude/-preservation behavior into plugin-local lib copy. |
| plugins/sync-docs/lib/adapter-transforms.js | Sync OpenCode .claude/-preservation behavior into plugin-local lib copy. |
| docs/consult-command-test-strategy.md | Update documented expectations for new model mappings. |
| README.md | Update default Gemini model name in top-level docs table. |
| tests/consult-command.test.js | Update tests to reflect canonical redaction ownership in consult SKILL.md. |
| tests/debate-command.test.js | Add comprehensive tests for debate plugin markdown structure and consistency. |
| tests/gen-adapters.test.js | Update tests to require frontmatter on line 1 (no header). |
| adapters/opencode/agents/agnix-agent.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/agent-enhancer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/ci-fixer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/ci-monitor.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/claudemd-enhancer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/consult-agent.md | Regenerated OpenCode adapter output (header removal + consult agent changes). |
| adapters/opencode/agents/cross-file-enhancer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/debate-orchestrator.md | Regenerated OpenCode adapter output (header removal + debate orchestrator changes). |
| adapters/opencode/agents/delivery-validator.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/deslop-agent.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/docs-enhancer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/exploration-agent.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/hooks-enhancer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/implementation-agent.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/learn-agent.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/map-validator.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/perf-analyzer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/perf-code-paths.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/perf-investigation-logger.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/perf-orchestrator.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/perf-theory-gatherer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/perf-theory-tester.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/plan-synthesizer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/planning-agent.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/plugin-enhancer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/prompt-enhancer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/simple-fixer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/skills-enhancer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/sync-docs-agent.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/task-discoverer.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/test-coverage-checker.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/agents/worktree-manager.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/agnix.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/audit-project-agents.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/audit-project-github.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/audit-project.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/consult.md | Regenerated OpenCode adapter output (header removal + consult command changes). |
| adapters/opencode/commands/debate.md | Regenerated OpenCode adapter output (header removal + debate command changes). |
| adapters/opencode/commands/delivery-approval.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/deslop.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/drift-detect.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/enhance.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/learn.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/next-task.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/perf.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/repo-map.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/ship-ci-review-loop.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/ship-deployment.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/ship-error-handling.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/ship.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/commands/sync-docs.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/agnix/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/consult/SKILL.md | Regenerated OpenCode adapter output (header removal + consult SKILL updates). |
| adapters/opencode/skills/debate/SKILL.md | Regenerated OpenCode adapter output (header removal + debate SKILL updates). |
| adapters/opencode/skills/deslop/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/discover-tasks/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/drift-analysis/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/enhance-agent-prompts/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/enhance-claude-memory/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/enhance-cross-file/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/enhance-docs/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/enhance-hooks/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/enhance-orchestrator/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/enhance-plugins/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/enhance-prompts/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/enhance-skills/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/learn/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/orchestrate-review/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/perf-analyzer/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/perf-baseline-manager/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/perf-benchmarker/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/perf-code-paths/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/perf-investigation-logger/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/perf-profiler/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/perf-theory-gatherer/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/perf-theory-tester/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/repo-mapping/SKILL.md | Regenerated OpenCode adapter output (header removal + .claude/ preserved in enumerations). |
| adapters/opencode/skills/sync-docs/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/opencode/skills/validate-delivery/SKILL.md | Regenerated OpenCode adapter output (header removal). |
| adapters/codex/skills/agnix/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/audit-project-agents/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/audit-project-github/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/audit-project/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/consult/SKILL.md | Regenerated Codex adapter output (header removal + consult command changes). |
| adapters/codex/skills/debate/SKILL.md | Regenerated Codex adapter output (header removal + debate command changes). |
| adapters/codex/skills/delivery-approval/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/deslop/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/drift-detect/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/enhance/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/learn/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/next-task/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/perf/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/repo-map/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/ship-ci-review-loop/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/ship-deployment/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/ship-error-handling/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/ship/SKILL.md | Regenerated Codex adapter output (header removal). |
| adapters/codex/skills/sync-docs/SKILL.md | Regenerated Codex adapter output (header removal). |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Comment on lines
+131
to
+146
| If context resolved to "file": | ||
| Use a follow-up AskUserQuestion to ask for the file path: | ||
| ``` | ||
| AskUserQuestion: | ||
| questions: | ||
| - header: "File path" | ||
| question: "Which file should both tools see?" | ||
| multiSelect: false | ||
| options: | ||
| - label: "src/" description: "Source directory file" | ||
| - label: "README.md" description: "Project readme" | ||
| ``` | ||
| The user can type any path via "Other". | ||
| After getting the path, validate it exists using the Read tool. | ||
| If the file doesn't exist: `[ERROR] Context file not found: {PATH}` | ||
| If valid, set context to `file={user_provided_path}`. |
There was a problem hiding this comment.
The follow-up file-path validation instructs using the Read tool on a user-provided path, but it doesn’t constrain the path to the project directory. That allows reading arbitrary absolute paths / path traversal during “validation” (before consult’s own context guards). Add an explicit safety check (reject absolute paths and any normalized path escaping cwd) before any Read, or delegate existence/validation entirely to consult’s --context=file=… handling without reading the file here.
Collaborator
Author
There was a problem hiding this comment.
Fixed - added path containment checks (reject absolute paths outside cwd, reject .., reject UNC paths) before the Read validation.
Comment on lines
103
to
106
| ### Round 2+: Challenger Follow-up | ||
|
|
||
| ``` | ||
| You are the CHALLENGER in round {round} of a structured debate. | ||
|
|
||
| Topic: {topic} | ||
|
|
||
| {context_summary} | ||
|
|
||
| The PROPOSER ({proposer_tool}) responded to your challenges: | ||
|
|
||
| --- | ||
| {proposer_previous_response} | ||
| --- | ||
|
|
||
| Your job: Evaluate the proposer's defense. For each point they addressed: | ||
| - Did they adequately address your concern? If so, acknowledge it | ||
| - Did they dodge or superficially address it? Call it out specifically | ||
| - Are there NEW weaknesses in their revised position? | ||
|
|
||
| If you're genuinely convinced on a point, say so - but explain what convinced you. | ||
| If you see new problems, raise them. | ||
|
|
||
| Provide your follow-up: | ||
| ``` | ||
| *(JavaScript reference - not executable in OpenCode)* | ||
|
|
There was a problem hiding this comment.
The OpenCode adapter lost the entire “Round 2+: Challenger Follow-up” prompt template: it was replaced with the generic *(JavaScript reference - not executable in OpenCode)* placeholder. This removes required debate behavior/instructions for OpenCode. Adjust the OpenCode adapter transform heuristics so non-JS prompt-template code fences like this one are preserved (e.g., require an explicit JS language tag, or tighten the keyword detection), then regenerate adapters.
Collaborator
Author
There was a problem hiding this comment.
Pre-existing adapter transform behavior - the OpenCode transformer strips code blocks that look like executable content. This affects all skills with template code blocks, not specific to this PR. The source skill has the full templates.
Comment on lines
+114
to
120
| - header: "Context" # SKIP if --context resolved | ||
| question: "Include codebase context for both tools?" | ||
| options: | ||
| - label: "None (Recommended)" description: "No extra context, just the topic" | ||
| - label: "Diff" description: "Include current git diff" | ||
| - label: "File" description: "Include a specific file (will ask path)" | ||
| ``` |
There was a problem hiding this comment.
The newly added Codex request_user_input example for the Context picker still doesn’t include the required per-question id fields (and the current in-block quoted note isn’t valid YAML for the tool call). Please update this example so each question (including the new Context question) has a unique id, and remove/replace the quoted note with valid fields.
Collaborator
Author
There was a problem hiding this comment.
Pre-existing Codex adapter format limitation. The id fields are injected by the adapter transform at install time, not in the source markdown.
Comment on lines
+342
to
+347
| for (const [filePath, content] of files) { | ||
| if (content.includes('---\n')) { | ||
| // Files with frontmatter must start with --- on line 1 | ||
| // (no auto-generated header before frontmatter) | ||
| expect(content.startsWith('---\n')).toBe(true); | ||
| } |
There was a problem hiding this comment.
content.includes('---\n') is not a reliable way to detect frontmatter, since many markdown bodies include --- as horizontal rules / separators. This test can produce false positives and fail on files that contain --- but don’t have YAML frontmatter. Consider detecting frontmatter with a start-of-file check (e.g., content.startsWith('---\n')) or a frontmatter-specific regex anchored to the beginning of the file.
Collaborator
Author
There was a problem hiding this comment.
Acknowledged. In practice all adapter files start with --- frontmatter (verified by gen-adapters output), but a more robust check using a regex anchor would be better. Low priority - the current test catches the actual regression it was written for (headers before frontmatter).
New sonnet is faster, cheaper, and close to opus capability. These agents do pattern matching and synthesis, not deep reasoning: - exploration-agent: reads code, maps patterns - learn-agent: web research + guide creation Keeps opus for: planning, implementation, perf-orchestrator, debate-orchestrator (judgment-intensive tasks).
Contributor
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 33 out of 33 changed files in this pull request and generated no new comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
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
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
Comprehensive fixes from running 3 debates + /enhance + 4-pass review on debate and consult plugins.
Debate plugin fixes
{round-1}->{previous_round}for clarityConsult plugin fixes
gpt-5-codex-mini/gpt-5.1-codex-mini->gpt-5.3-codex-spark--continue+--count > 1guard in agentclaude-opus-4-5->claude-opus-4-6Infrastructure
.claude/in platform documentation enumerationsTest plan
gpt-5-codex-miniorgpt-5.1-codex-miniremaining