Release hardening for v1.0.0: harness safety, honest docs, changelog#3
Merged
Merged
Conversation
Make the integration safe and accurate for real agent-harness use, and add release hygiene. - AGENTS.md: correct the Setup note (no rtk init hook for Command Code yet); add a "Harness safety" section — don't wrap streaming/-f output (RTK buffers and can hang), trust the raw exit code when a pass/fail verdict matters, and prefer native file/search tools over rtk ls/grep/find/read. - references/commands.md: drop the "near-lossless" overstatement; mark the log rows static-only (never -f/follow); add a "Harness notes" block (streaming, exit status, piped/NO_COLOR decoration, native tools). - SKILL.md: add a "Compatibility & limitations" section; bump to 1.0.0. - README.md: add a "Compatibility & limitations" section (Command Code hook gap, exit-code fidelity, piped/ANSI per RTK #1282, streaming, PATH, native tools, permission surface, Windows hook); make the savings table honest (large output only; small can be net-negative; static logs). - Add CHANGELOG.md (1.0.0). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VvpBXNCbbg4eibmZQZAbb6
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Final release-readiness pass. Makes the integration safe and accurate for real agent-harness use and adds release hygiene, so wrapping commands in
rtkwon't quietly cause tool failures or cost tokens.Hardening against harness-level failure modes (several are documented RTK issues):
okwhen a command failed) and propagation isn't guaranteed per command/version. Docs now say: for a pass/fail that matters (tests, gates), trust the raw exit code or run raw /rtk proxy.NO_COLOR=1noted.-f/tail -f/ growing logs can hang. Marked "run raw"; log rows tagged static-only.rtk; treated as optional with a raw fallback (no-op, not a failure).rtk read/grep/find.rtk/rtk proxycan execute arbitrary wrapped commands; flagged as a broad allow-list entry.rtk initdoesn't target Command Code, so manual prefixing is the working default.Docs / release hygiene:
README.mdandSKILL.md(with the above + RTK issue links).AGENTS.mdand Harness notes inreferences/commands.md.Type of change
references/commands.md)install.sh/install.ps1)README.md/SKILL.md/AGENTS.md)Checklist
*.ps1)SKILL.mdstill has valid frontmatter (name:,description:)SKILL.md/AGENTS.mdvalidation — verified locally)rtkcommand/flag referenced is confirmed against the real rtk-ai/rtk command surface🤖 Generated with Claude Code
https://claude.ai/code/session_01VvpBXNCbbg4eibmZQZAbb6
Generated by Claude Code