Skip to content

fix: unify concise built-in prompt defaults - #705

Merged
LeXwDeX merged 2 commits into
mainfrom
fix/concise-global-prompts
Oct 3, 2026
Merged

LeXwDeX merged 2 commits into
mainfrom
fix/concise-global-prompts

Conversation

@LeXwDeX

@LeXwDeX LeXwDeX commented Oct 3, 2026 •

Copy link
Copy Markdown
Owner

Why

Built-in prompts used inconsistent writing rules. Generic one-word and fixed-line reply limits also conflicted with clear explanations.

What changed

This change adds seven clear-writing defaults to both conversation runtimes. Users do not need to edit configuration. Explicit user instructions and task-specific formats take precedence.

It also simplifies Memory, reasoning distillation, model, agent, goal, command, capability, and tool prompts. It preserves data schemas, source and evidence checks, user confirmation, permissions, budgets, and output formats. Recorded native requests now include the new defaults; recorded responses and matching rules are unchanged.

Stable release notes are provided for GraphAgent v1.0.61.

Evidence

  • Memory and reasoning distillation: 335 tests passed.
  • Goal prompts and judge: 35 tests passed.
  • Global prompt assembly: 14 tests passed.
  • Native recorded tool loops: 3 tests passed; 1 skipped because its cassette is absent.
  • Hook event wiring and DAG output contract: 32 tests passed. Required contract phrases remain intact.
  • Workspace typecheck: 31 tasks passed.
  • Lint: 4,832 warnings, 0 errors; the existing 4,850 limit is unchanged.
  • Local CLI test build completed. Isolated Qwen, GLM, and DeepSeek calls kept the source/backup paths and pending integrity check. All returned text with no tool calls. The test build ran directly because the administrator-owned installation requires a password.
  • Astra final code review passed. Native CI remains the merge gate.

Checklist

  • Add defaults to both runtimes and preserve explicit output formats.
  • Simplify built-in prompts and retain Memory/distillation contracts.
  • Complete local checks, isolated model calls, and Astra final review.
  • Pass current-head native CI before merge.
  • Confirm merge and both Issue closures before stable publication.

Closes #703
Closes #704

Native CI passed at f50bba911815f6258a7dc3655a1408cff3e15686: Typecheck, complete unit and Linux/Windows E2E gates. The opencode package finished with 5,158 passing tests, 33 skips, and zero failures. Astra approved the final code and publication plan.

Delivery complete: merged into main at 39c8067d664e1327d67b8df9c9709f39e268959f; #703 and #704 are closed. The merged tree matches the reviewed PR head. Native main Typecheck and complete test gates passed.

Stable v1.0.61 is published as Latest from that merge commit. The release workflow passed for Linux, macOS and Windows. All six assets are present; SHA256SUMS matches GitHub asset digests and template provenance points to the merge commit. SHA256SUMS, provenance and the macOS archive were downloaded and verified; the packaged macOS CLI passed version 1.0.61, help and installer-style signing checks. Local administrator-owned installation remains at 1.0.60.

Apply clear writing defaults without configuration. Simplify Memory, reasoning distillation and built-in prompts while preserving contracts. Prepare the 1.0.61 stable release.
@LeXwDeX
LeXwDeX marked this pull request as ready for review October 3, 2026 07:51
@chatgpt-codex-connector

Copy link
Copy Markdown

The account paying for this security review has reached its Codex usage limits. The payer can check the Codex usage dashboard. For personal accounts, using credits requires enabling “Use credits for security reviews” in Code review settings. If you do not manage the paying account, contact this repository's admins.

@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Oct 3, 2026 •

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-10-03T08:18:29.636056Z f50bba9 New commits
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@chatgpt-codex-connector

Copy link
Copy Markdown

The account paying for this security review has reached its Codex usage limits. The payer can check the Codex usage dashboard. For personal accounts, using credits requires enabling “Use credits for security reviews” in Code review settings. If you do not manage the paying account, contact this repository's admins.

@LeXwDeX
LeXwDeX merged commit 39c8067 into main Oct 3, 2026
9 checks passed
@LeXwDeX
LeXwDeX deleted the fix/concise-global-prompts branch October 3, 2026 08:55
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

fix: simplify memory and reasoning distillation instructions fix: apply clear global writing defaults

1 participant