coldpass

Parcha-ai/agentrun

GitHub ↗ · How this check works

Per-setup public labels: Linux / Haiku: Works

Static status badge · SVG · read-only JSON. Embed with ![Coldpass status](https://coldpass.dev/badge/Parcha-ai--agentrun.svg).

Works

Only older checks are saved; no current product test.

Answers apply to each recorded setup, revision and operation. Earlier observations remain in Run details.

Tested on Linux
PlatformHaikuLuna
Linuxyes

Linux / Haiku 5.5: yes · Oct 17, 2026 · 7202ae494293 · claude-code / 2.1.300 (Claude Code)
Recorded operation: /opt/node22/bin/node packages/dsl/dist/cli.js demo

Needs keys or browser
None seen
Setup time
2 s to install, 1 min 59 s to first use
Issue
none found in recorded check

Safety: no observations were recorded for the selected revision. Functional results do not establish a safety result.

Repair history and test corrections
Current diagnosis: unknown · resolved

Coldpass condensed instruction text introduced newline controls rejected by the existing job schema before any model call. Job construction now normalizes instructional whitespace once, including repair context; exact original failure remains immutable. Default/repair schema regression passes.

Repair record fix-e20faf7e09b039cc3bf4 · current status JSON. Original run outcomes remain unchanged.

Failed run: Parcha-ai/agentrun · Oct 16, 2026 · Our test failed; no repo verdict.

Test-repair rerun: Parcha-ai/agentrun · Oct 16, 2026 · Not tested.

Run details

Haiku 5.5

Native agent · Checked operation passed

Checked operation /opt/node22/bin/node packages/dsl/dist/cli.js demo

Agent-reported scope: Ran the README quickstart demo `agentrun demo` (packages/dsl/dist/cli.js demo): a scripted ticket-routing workflow; it completed with Status: complete and queue=billing.

Run details
run
Parcha-ai--agentrun-claude-haiku-5-5-1792231078918725969
date
Oct 17, 2026
time to working
145.4s
product hosting / execution
Local command in the guest
recorded operation scope
product operation
model
claude-haiku-5-5
agent / version
claude-code / 2.1.300 (Claude Code)
platform
linux
reported cost
$0.005547
Recorded source
https://github.com/Parcha-ai/agentrun
Tool transport
mcp-assigned-vm
Test version
readme-520bfd15730ba11f12f732859d1d8be557b2f38c2b58468f19f590c551dabe62
revision
7202ae49429393d41e744131e75105c03cb7d99d
Whose fault: not assigned · resolved

What the agent did

Ran the README quickstart demo `agentrun demo` (packages/dsl/dist/cli.js demo): a scripted ticket-routing workflow; it completed with Status: complete and queue=billing.

apt-get install -y xz-utils (needed to unpack the Node tarball; apt default node is 18.19, below required 22.19+)
curl -sfL https://nodejs.org/dist/v22.23.2/node-v22.23.2-linux-x64.tar.xz | tar -xJ -C /opt/node22 --strip-components=1 (matches .nvmrc 22.23.2)
npm ci --ignore-scripts
npm run build
node packages/dsl/dist/cli.js demo

run JSON · checker evidence · agent report + check · transcript excerpt · error excerpt

Earlier attempts (1)

Haiku 5.5

Native agent · Not tested

Checked operation npm run demo:support -- payment

Agent-reported scope: Ran the README's no-key support workflow demo: npm run demo:support -- payment (scripted tools, Jev answers and agent responses; no live calls). Result: status complete, 1 agent call, 2 decision calls, returned the investigation answer, matching the README table row 'Investigate a failed payment'.

Run details
run
Parcha-ai--agentrun-claude-haiku-5-5-1792207734310717294
date
Oct 16, 2026
time to working
147.0s
product hosting / execution
Local command in the guest
recorded operation scope
product operation
model
claude-haiku-5-5
agent / version
claude-code / 2.1.300 (Claude Code)
platform
linux
reported cost
$0.005129
Recorded source
https://github.com/Parcha-ai/agentrun
Tool transport
mcp-assigned-vm
Test version
readme-d7385db4c550d6450cd4f7a027673c8ef032179188dfdf5561cf0652fc6e888a
revision
7202ae49429393d41e744131e75105c03cb7d99d
Current diagnosis: unknown · resolved

Coldpass condensed instruction text introduced newline controls rejected by the existing job schema before any model call. Job construction now normalizes instructional whitespace once, including repair context; exact original failure remains immutable. Default/repair schema regression passes.

Repair record fix-e20faf7e09b039cc3bf4 · current status JSON. Original run outcomes remain unchanged.

Failed run: Parcha-ai/agentrun · Oct 16, 2026 · Our test failed; no repo verdict.

Test-repair rerun: Parcha-ai/agentrun · Oct 16, 2026 · Not tested.

Double-checks failed run: Parcha-ai/agentrun · Oct 16, 2026.

Fix: fix-e20faf7e09b039cc3bf4.

What the agent did

Ran the README's no-key support workflow demo: npm run demo:support -- payment (scripted tools, Jev answers and agent responses; no live calls). Result: status complete, 1 agent call, 2 decision calls, returned the investigation answer, matching the README table row 'Investigate a failed payment'.

Installed Node v22.23.2 from https://nodejs.org/dist/v22.23.2/node-v22.23.2-linux-x64.tar.xz (required by README: Node 22.19+; apt offers only Node 18) after installing xz-utils
npm ci --ignore-scripts
npm run build
npm run demo:support -- payment

run JSON · checker evidence · agent report + check · transcript excerpt · error excerpt

No product verdict. Run record.