mcpx: shebang-first interpreter dispatch + script.interpreter + script.cwd (split of #1919) #2167

Activity

1. dezren39

   added

   enhancementNew feature or requestNew feature or request

   priority/normalOrdinary priority (the default)Ordinary priority (the default)

   queue/queuedHas a queued task in the Supabase agent queueHas a queued task in the Supabase agent queue

   app/mcpxPackage/app: mcpx (pkgs/ or apps/ in this repo)Package/app: mcpx (pkgs/ or apps/ in this repo)

   on Oct 29, 2026
2. developing-today commented on Oct 29, 2026

   

   developing-todayBot

   on Oct 29, 2026 – with developing.today

   Contributor

   More actions

   CODE-1060
3. github-actions

   assigned

   dezren39

   on Oct 29, 2026
4. dezren39

   added

   work/assignedQueue worker lifecycle stateQueue worker lifecycle state

   queue/parkedQueue task is parked (waiting on CI, time, or an event)Queue task is parked (waiting on CI, time, or an event)

   status/parkedDeliberately paused with a recorded resume conditionDeliberately paused with a recorded resume condition

   and removed

   queue/queuedHas a queued task in the Supabase agent queueHas a queued task in the Supabase agent queue

   on Oct 30, 2026
5. dezren39 commented on Oct 30, 2026

   

   dezren39

   on Oct 30, 2026

   MemberAuthor

   More actions

   🤖 Agent note · parked (not started, lane budget)

   • Agent: openchamber scheduled lane (queue worker repo-worker-code)
   • Session / ID: queue task 3201 (small_fix, mcpx: shebang-first interpreter dispatch + script.interpreter + script.cwd (split of #1919) #2167 parts 1-3 of mcpx: exec should run any shebang, and resolve file scripts the same way local or remote #1919) — claimed with ~12 min left in a 30-min lane, parked 20 min, no branch created
   • Why not started: this is a 45-minute body of work (dispatch order + script.interpreter in the settings registry + script.cwd policy + non-JS env/exec handling + tests). Two earlier tasks in this shift were blocked on the repo-wide CI runner stall, which left no honest window for it; a half-built branch would be worse than none.
   • Recon for the next lane: the packet cites runner.Detect returning three JS runtimes — that symbol does not exist on main 3a425552. internal/runner/runtimes.go exposes ParseSetup, Setup.lookup, Resolve(prefer, permSpec, s) and build(candidate, perms); there is no per-language dispatch function there. Start by finding the actual dispatch site (grep for the configured TypeScript runtime across internal/cli, internal/execsvc, internal/api) before implementing, and treat the packet's file:line evidence as stale until confirmed.
   • Also: the repo's Actions pool was stalled all shift (24 runs queued, no non-skipped job completing since ~05:10Z), so any PR opened here would have parked on CI too.
   • Resume: confirm the dispatch site → implement shebang > script.interpreter > script.language/extension > configured TS runtime → script.cwd → tests.
6. dezren39

   added

   queue/claimedQueue worker lifecycle stateQueue worker lifecycle state

   status/in-progressQueue worker lifecycle stateQueue worker lifecycle state

   and removed

   status/parkedDeliberately paused with a recorded resume conditionDeliberately paused with a recorded resume condition

   queue/parkedQueue task is parked (waiting on CI, time, or an event)Queue task is parked (waiting on CI, time, or an event)

   on Oct 30, 2026
7. 235 remaining items

   Load more

   Load more actions
8. dezren39

   added

   queue/parkedQueue task is parked (waiting on CI, time, or an event)Queue task is parked (waiting on CI, time, or an event)

   status/parkedDeliberately paused with a recorded resume conditionDeliberately paused with a recorded resume condition

   and removed

   queue/claimedQueue worker lifecycle stateQueue worker lifecycle state

   on Oct 31, 2026
9. dezren39 commented on Oct 31, 2026

   

   dezren39

   on Oct 31, 2026

   MemberAuthor

   More actions

   agent-note:v1
   event: parked
   task: 3201 (repo-health-code-pilot, lane B)
   state: Parked 20m on CI. Run 41211362870: all jobs complete except final: CI time budget (queued, runner backlog). stress: PR changed tests FAILED but its log is unavailable until the run completes, so the failing tests are not yet identified. Codex clean on 0847133; lint, plugin tests, docs build pass; laptop post-merge execsvc/runner/daemon PASS (lane F).
   resume: run 41211362870 complete -> read stress job 124566873865 log -> if failures are only the filed main-baseline family (#2846/#2716/#2697) or the proven one-off flake, merge -> close #2167 -> complete fixed_merged; if a PR-touched test failed, fix on branch queue/3201-2167-shebang-dispatch first.
10. dezren39

    added

    queue/claimedQueue worker lifecycle stateQueue worker lifecycle state

    status/in-progressQueue worker lifecycle stateQueue worker lifecycle state

    and removed

    status/parkedDeliberately paused with a recorded resume conditionDeliberately paused with a recorded resume condition

    queue/parkedQueue task is parked (waiting on CI, time, or an event)Queue task is parked (waiting on CI, time, or an event)

    on Oct 31, 2026
11. dezren39 commented on Oct 31, 2026

    

    dezren39

    on Oct 31, 2026

    MemberAuthor

    More actions

    🤖 Agent note · picked up

    • Agent: queue worker (repo-worker-code), lane E shift
    • Session / ID: queue task 3201
    • Found via: queue claim
    • Task: Resuming from awaiting_ci — identify the stress-job failures on CI run 41211362870 for PR mcpx: shebang-first interpreter dispatch + script.interpreter + script.cwd (#2167) #2611 (head 0847133, Codex clean per prior checkpoint); merge only if failures are the filed pre-existing baseline family.
    • Expectation: effort s — done = PR merged + issue closed, or task parked with the exact remaining gate named.
    • Links: PR mcpx: shebang-first interpreter dispatch + script.interpreter + script.cwd (#2167) #2611 · issue mcpx: shebang-first interpreter dispatch + script.interpreter + script.cwd (split of #1919) #2167
12. dezren39

    added a commit that references this issue on Oct 31, 2026

    mcpx: shebang-first interpreter dispatch + script.interpreter + scrip…

    ...

    8e7b504
13. dezren39 commented on Oct 31, 2026

    

    dezren39

    on Oct 31, 2026

    MemberAuthor

    More actions

    🤖 Agent note · done

    • Agent: queue worker (repo-worker-code), lane E shift
    • Session / ID: queue task 3201
    • Found via: queue claim (resume from awaiting_ci)
    • Task: PR mcpx: shebang-first interpreter dispatch + script.interpreter + script.cwd (#2167) #2611 merged — squash merge commit 8e7b504, head 0847133.
    • Evidence: Stress job log for CI run 41211362870 (fetched per-job): the PR's own packages pass under race/shuffle/repeat — runner ok 25.8s, execsvc ok 30.8s, daemon ok 275.8s; lint, plugin tests, and nix docs build pass. The only failures are filed clean-main baseline: the four internal/settings guard tests (mcpx: internal/settings tests red on clean main — TESTING.md missing from compendium manifest, model.* settings lack effect probes, env doc drift #2846 TESTING.md compendium + model.* probes, mcpx: internal/settings (scriptenv) and internal/conformance (TestAuthSplit, TestMatrix) failing on clean main #2716 scriptenv, mcpx: three internal/settings guard tests fail on main (compendium chapter + undeclared adapter env) #2697 adapter env MCPX_ADAPTER_DECLARATION_SHA256 — none of the named settings/env/files are this PR's; script.interpreter/script.cwd have effect probes and are absent from the violation lists) and the internal/cli 600s stress package timeout (mcpx: internal/cli package test duration exceeds the 600s stress timeout on PR 2714 (stage 1 of #1980) #2859, same failure filed from PR 2714; the timed-out goroutine was runnable inside pre-existing completion generation, completion.go untouched by this PR). Codex: zero unresolved review threads on the final head. The run's final aggregator job remains queued in the runner backlog and can only reflect the already-analysed stress result.
    • Expectation: done — closing this issue; baseline failures stay tracked on mcpx: internal/settings tests red on clean main — TESTING.md missing from compendium manifest, model.* settings lack effect probes, env doc drift #2846/mcpx: internal/settings (scriptenv) and internal/conformance (TestAuthSplit, TestMatrix) failing on clean main #2716/mcpx: three internal/settings guard tests fail on main (compendium chapter + undeclared adapter env) #2697/mcpx: internal/cli package test duration exceeds the 600s stress timeout on PR 2714 (stage 1 of #1980) #2859.
    • Links: PR mcpx: shebang-first interpreter dispatch + script.interpreter + script.cwd (#2167) #2611 (merged) · queue task 3201
14. dezren39

    removed

    status/in-progressQueue worker lifecycle stateQueue worker lifecycle state

    queue/claimedQueue worker lifecycle stateQueue worker lifecycle state

    work/assignedQueue worker lifecycle stateQueue worker lifecycle state

    on Oct 31, 2026
15. dezren39

    closed this as completedon Oct 31, 2026