star-chamber: raise anthropic max_tokens in reference config to match peers

[peteski22]

What

Raise anthropic's max_tokens from 8192 → 65536 in the bundled reference provider config (plugins/pragma/reference/star-chamber/providers.json).

Why

The reference config that seeds a user's star-chamber provider config had an inconsistent per-provider token skew:

provider	before	after
openai	128000	128000
anthropic	8192	65536
gemini	65536	65536

Opus emits long, structured JSON reviews. An 8192 cap truncates the response mid-JSON, so the council can't extract valid JSON and silently drops the anthropic judge — a council that should have three judges runs with two. This is a config default, not a model or parser limitation.

65536 matches gemini, so a config seeded from this template produces a valid three-judge council out of the box.

Scope

One-line value change. No schema, code, or doc changes required — PROTOCOL.md's generic max_tokens default (16384) and examples are unrelated to this per-provider template.

Verification

• python3 -m json.tool confirms the file is still valid JSON.
• Confirmed no other 8192 occurrences remain in plugins/.

Summary by CodeRabbit

• Improvements
  • Increased the maximum token limit for the Anthropic provider from 8,192 to 65,536, enabling support for substantially longer requests or responses.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: aca3b631-65dd-40a6-b49f-c47d75a477af

📥 Commits

Reviewing files that changed from the base of the PR and between bc7c0cc and f8aaf05.

📒 Files selected for processing (1)

• plugins/pragma/reference/star-chamber/providers.json

---

📝 Walkthrough

Walkthrough

The Anthropic provider configuration increases its max_tokens limit from 8192 to 65536.

Changes

Provider configuration

Layer / File(s)	Summary
Anthropic token limit plugins/pragma/reference/star-chamber/providers.json	The Anthropic provider now permits up to 65536 tokens instead of 8192.

Estimated code review effort: 1 (Trivial) | ~2 minutes

Suggested reviewers: copilot

Poem

A rabbit checks the token gate,
And finds the ceiling raised in state.
Eight thousand hops are now no more,
Sixty-five thousand mark the door.
More room for words to gently flow.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly and concisely summarises the increase to Anthropic's max_tokens limit in the reference configuration.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/star-chamber-anthropic-max-tokens

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[Copilot]

Pull request overview

Raises the Anthropic provider max_tokens limit in the bundled star-chamber reference configuration so that councils seeded from this template are less likely to truncate long, structured (JSON) review outputs.

Changes:

• Increased anthropic.max_tokens from 8192 to 65536 in the reference providers.json template.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.