docs: verify the herdr flow works end to end#31
Closed
jakemassoth wants to merge 1 commit into
Closed
Conversation
QA/verification ticket (not a feature): exercised the full herdr-backed lifecycle — init, project add/list, refresh, spawn, peek, send, status, ticket status-set, restart, kill, cleanup — end to end against a real herdr 0.7.4 server, plus the mock-backed unit suite. Verified at three levels: 55 unit tests, 19 real-herdr e2e tests, and a hands-on isolated walkthrough capturing real command output + pane text. All isolated on a throwaway herdr session + private XDG_CONFIG_HOME + stand-in agents; the live 'yeschef'/'default' brigade was never touched and no sandbox leaked. No breakage found; two cosmetic, non-blocking observations recorded (unknown agent_status for stand-in agents; status AGENT-column overflow with long --agent strings).
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
QA/verification ticket — prove the whole new herdr flow works, end to end (herdr
replaced tmux as yeschef's session/TUI backend in #29). Not a feature; the deliverable
is a verification report, added as
docs/herdr-flow-verification.md. No source changes.Verdict: VERIFIED ✅
Every stage of the lifecycle yeschef exposes on top of herdr works end to end against a
real
herdr 0.7.4server:init,project add/list,refresh,spawn,peek,send,status,ticket status-set,restart,kill,cleanup. Confirmed at threelevels:
cargo test --bin yeschef— pass.nix run .#e2e— pass (5.5s).real command output + pane text (spawn → peek shows the agent's markers → send is
echoed back → status/status-set → restart restores the brigade → kill → cleanup reaps).
Isolation — the live kitchen was never touched
A real brigade (
yeschef+defaultsessions) was running the whole time. All herdr opswere confined to a sandbox: a unique session name, a private
XDG_CONFIG_HOMEunder
/tmp, a privateYESCHEF_HOME, and stand-in agents only (sh/cat/sleep— no real coding agent, no network). Server + sandbox torn down on exit.Post-run: both live sessions still
running, no leaked session, no leftover sandbox.Observations (non-blocking, not breakage)
agent_statusreadsunknownfor stand-in agents — herdr only detects real agentintegrations, so herdr's own
working/idle/blocked/donedetection and nativeconversation-resume-on-restart were not re-proven (they need a live recognized agent,
deliberately out of scope). Everything yeschef itself drives is fully verified.
statusAGENT column overflows its fixed width with a long--agentstring (cosmetic;the default
claudefits). Left as-is; candidate for a future polish.See
docs/herdr-flow-verification.mdfor the full report, evidence, and reproduce steps.