feat: add health snapshot reporting + CI artifacts by GopiB9119 · Pull Request #3 · GopiB9119/agent-live-web

GopiB9119 · 2026-03-05T10:35:10Z

Summary

Add scripts/health-snapshot.js to run JS syntax checks + Python agent unit tests and write timestamped reports.
Add npm scripts:
- health:snapshot
- health:snapshot:quiet
Extend python-agent-tests workflow to:
- setup Node + install npm deps
- run health snapshot in quiet mode
- upload logs/health/*.md artifacts
Update README with usage and retention behavior.

Validation

node --check scripts/health-snapshot.js
cmd /c npm run health:snapshot:quiet
npm run agent:test:py

Copilot

Pull request overview

Adds a new “health snapshot” script to generate timestamped health reports (JS syntax checks + Python agent unit tests), wires it into npm scripts, and extends CI workflows to run/report these checks and adjust evaluation gating behavior.

Changes:

Add scripts/health-snapshot.js to run Node --check across key JS files + Python unit tests and emit logs/health/health-snapshot-*.md reports with retention pruning.
Add npm scripts health:snapshot and health:snapshot:quiet.
Update CI: run the health snapshot in python-agent-tests and upload the generated markdown reports as artifacts; refine eval-gate PR vs manual gating.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 6 comments.

Show a summary per file

File	Description
scripts/health-snapshot.js	New report generator for JS syntax + Python unit test health snapshots with pruning.
package.json	Adds npm scripts to run the health snapshot (normal/quiet).
README.md	Documents health snapshot usage and report retention behavior.
.github/workflows/python-agent-tests.yml	Adds Node setup + health snapshot + artifact upload; changes workflow triggers/concurrency.
.github/workflows/eval-gate.yml	Adds branch scoping + concurrency; splits PR baseline gate from manual strict gate.

Copilot · 2026-03-05T10:41:13Z

+  if (process.env.AGENT_PYTHON && String(process.env.AGENT_PYTHON).trim()) {
+    candidates.push({ cmd: String(process.env.AGENT_PYTHON).trim(), argsPrefix: [] });
+  }
+
+  const winVenvPython = path.resolve(process.cwd(), '.venv', 'Scripts', 'python.exe');
+  if (fs.existsSync(winVenvPython)) {
+    candidates.push({ cmd: winVenvPython, argsPrefix: [] });
+  }
+


pythonCommandCandidates() only checks a Windows venv path and then falls back to python/py -3. On Linux/macOS it’s common to have .venv/bin/python and/or only python3 available, so health:snapshot can fail locally even when a venv exists. Add POSIX venv + python3 candidates (and keep the existing Windows/AGENT_PYTHON logic).

Suggested change

if (process.env.AGENT_PYTHON && String(process.env.AGENT_PYTHON).trim()) {

candidates.push({ cmd: String(process.env.AGENT_PYTHON).trim(), argsPrefix: [] });

}

const winVenvPython = path.resolve(process.cwd(), '.venv', 'Scripts', 'python.exe');

if (fs.existsSync(winVenvPython)) {

candidates.push({ cmd: winVenvPython, argsPrefix: [] });

}

const cwd = process.cwd();

if (process.env.AGENT_PYTHON && String(process.env.AGENT_PYTHON).trim()) {

candidates.push({ cmd: String(process.env.AGENT_PYTHON).trim(), argsPrefix: [] });

}

if (process.platform === 'win32') {

const winVenvPython = path.resolve(cwd, '.venv', 'Scripts', 'python.exe');

if (fs.existsSync(winVenvPython)) {

candidates.push({ cmd: winVenvPython, argsPrefix: [] });

}

} else {

const posixVenvPython = path.resolve(cwd, '.venv', 'bin', 'python');

if (fs.existsSync(posixVenvPython)) {

candidates.push({ cmd: posixVenvPython, argsPrefix: [] });

}

const posixVenvPython3 = path.resolve(cwd, '.venv', 'bin', 'python3');

if (fs.existsSync(posixVenvPython3)) {

candidates.push({ cmd: posixVenvPython3, argsPrefix: [] });

}

}

candidates.push({ cmd: 'python3', argsPrefix: [] });

Copilot · 2026-03-05T10:41:13Z

+    child.on('close', (code) => {
+      resolve({
+        ok: code === 0,
+        code: code || 0,
+        stdout,
+        stderr,
+        startedAt,
+        finishedAt: new Date(),
+        command: [command, ...args].join(' ')


In the close handler, code can be null when the process exits due to a signal. Using code: code || 0 will record an exit code of 0 even though ok is false, which makes the report misleading. Capture the (code, signal) parameters and set a non-zero code when code is null (and ideally include the signal in the recorded error output/metadata).

Suggested change

child.on('close', (code) => {

resolve({

ok: code === 0,

code: code || 0,

stdout,

stderr,

startedAt,

finishedAt: new Date(),

command: [command, ...args].join(' ')

child.on('close', (code, signal) => {

const finishedAt = new Date();

const exitCode = typeof code === 'number' ? code : 1;

let finalStderr = stderr;

if (signal) {

const signalMessage = `Process terminated by signal ${signal}`;

finalStderr = finalStderr ? `${finalStderr}\n${signalMessage}` : signalMessage;

}

resolve({

ok: exitCode === 0,

code: exitCode,

stdout,

stderr: finalStderr,

startedAt,

finishedAt,

command: [command, ...args].join(' '),

signal: signal || null

Copilot · 2026-03-05T10:41:13Z

+  for (const file of files) {
+    const result = await runCommand(process.execPath, ['--check', file], { streamOutput: !IS_QUIET });
+    outputs.push(`$ node --check ${file}`);
+    if (result.stdout) outputs.push(result.stdout.trim());
+    if (result.stderr) outputs.push(result.stderr.trim());
+
+    if (!result.ok) {
+      return {
+        ok: false,
+        code: result.code,
+        stdout: outputs.filter(Boolean).join('\n\n'),
+        stderr: result.stderr || '',
+        startedAt,
+        finishedAt: new Date(),
+        command: 'node --check <repo-js-files>'
+      };


runJsCheck() appends result.stderr into stdout (via outputs) and also returns it separately as stderr. Since formatSection() concatenates stdout and stderr, JS check failures will duplicate the same stderr content in the report. Return either a single combined output, or keep stdout/stderr separate (but don’t include stderr in both).

Copilot · 2026-03-05T10:41:14Z

 on:
  push:
-    paths:
-      - "agent/**"
-      - ".github/workflows/python-agent-tests.yml"
+    branches:
+      - main
  pull_request:
-    paths:
-      - "agent/**"
-      - ".github/workflows/python-agent-tests.yml"
+    branches:
+      - main


This workflow no longer uses paths: filters and will run for every PR targeting main, even when the Python agent isn’t touched. If the intent is to keep this scoped to agent/** (as the README currently states), consider restoring paths: (possibly in addition to branches:) to avoid unnecessary CI time.

Copilot · 2026-03-05T10:41:14Z

+npm run health:snapshot
+npm run health:snapshot:quiet
 ```

+`npm run health:snapshot` runs JS syntax checks plus Python agent unit tests and writes a timestamped report to `logs/health/`.
+Use `npm run health:snapshot:quiet` for CI-friendly minimal console output while preserving full report details.
+Report retention is automatic: by default the latest `20` reports are kept in `logs/health/`. Override with `HEALTH_SNAPSHOT_KEEP_LATEST`.


README says GitHub Actions runs the Python agent tests automatically for agent/** changes, but python-agent-tests.yml now triggers on all PRs to main (no paths: filter). Update the README statement to match the new workflow behavior, or re-scope the workflow to agent/** changes.

Copilot · 2026-03-05T10:41:14Z

+          npm run eval:collect
+          npm run eval:run
+          npm run eval:history
+          node -e "const fs=require('fs');const r=JSON.parse(fs.readFileSync('eval/report.json','utf8'));const score=Number((r.summary||{}).overallScore||0);if(score<40){console.error('[EvalBaselineGate] FAIL overallScore',score,'< 40');process.exit(1)};console.log('[EvalBaselineGate] PASS overallScore='+score);"


The PR baseline gate uses an inline node -e ... snippet with a hard-coded threshold (< 40) to parse eval/report.json. This duplicates logic already implemented in scripts/eval/ci-gate.js (npm run eval:gate) and makes future changes harder. Consider reusing the existing gate script and setting EVAL_GATE_MIN_SCORE=40 (and any other needed env flags) for the PR baseline case instead of embedding custom parsing here.

Suggested change

node -e "const fs=require('fs');const r=JSON.parse(fs.readFileSync('eval/report.json','utf8'));const score=Number((r.summary||{}).overallScore||0);if(score<40){console.error('[EvalBaselineGate] FAIL overallScore',score,'< 40');process.exit(1)};console.log('[EvalBaselineGate] PASS overallScore='+score);"

npm run eval:gate

env:

EVAL_GATE_MIN_SCORE: '40'

Gopi B added 3 commits March 5, 2026 15:20

ci: require both checks on main pull requests

a4ffc90

ci: make eval check PR-safe and keep strict manual gate

4698325

feat: add health snapshot command and CI artifact upload

0d62d66

Copilot AI review requested due to automatic review settings March 5, 2026 10:35

Copilot started reviewing on behalf of GopiB9119 March 5, 2026 10:35 View session

GopiB9119 merged commit 212ba58 into main Mar 5, 2026
4 checks passed

GopiB9119 deleted the feat/health-snapshot-reporting branch March 5, 2026 10:37

Copilot AI reviewed Mar 5, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat: add health snapshot reporting + CI artifacts#3

feat: add health snapshot reporting + CI artifacts#3
GopiB9119 merged 3 commits into
mainfrom
feat/health-snapshot-reporting

GopiB9119 commented Mar 5, 2026

Uh oh!

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Mar 5, 2026

Uh oh!

Copilot AI Mar 5, 2026

Uh oh!

Copilot AI Mar 5, 2026

Uh oh!

Copilot AI Mar 5, 2026

Uh oh!

Copilot AI Mar 5, 2026

Uh oh!

Copilot AI Mar 5, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

-  if (process.env.AGENT_PYTHON && String(process.env.AGENT_PYTHON).trim()) {
-    candidates.push({ cmd: String(process.env.AGENT_PYTHON).trim(), argsPrefix: [] });
-  }
-  const winVenvPython = path.resolve(process.cwd(), '.venv', 'Scripts', 'python.exe');
-  if (fs.existsSync(winVenvPython)) {
-    candidates.push({ cmd: winVenvPython, argsPrefix: [] });
-  }
+  const cwd = process.cwd();
+  if (process.env.AGENT_PYTHON && String(process.env.AGENT_PYTHON).trim()) {
+    candidates.push({ cmd: String(process.env.AGENT_PYTHON).trim(), argsPrefix: [] });
+  }
+  if (process.platform === 'win32') {
+    const winVenvPython = path.resolve(cwd, '.venv', 'Scripts', 'python.exe');
+    if (fs.existsSync(winVenvPython)) {
+      candidates.push({ cmd: winVenvPython, argsPrefix: [] });
+    }
+  } else {
+    const posixVenvPython = path.resolve(cwd, '.venv', 'bin', 'python');
+    if (fs.existsSync(posixVenvPython)) {
+      candidates.push({ cmd: posixVenvPython, argsPrefix: [] });
+    }
+    const posixVenvPython3 = path.resolve(cwd, '.venv', 'bin', 'python3');
+    if (fs.existsSync(posixVenvPython3)) {
+      candidates.push({ cmd: posixVenvPython3, argsPrefix: [] });
+    }
+  }
+  candidates.push({ cmd: 'python3', argsPrefix: [] });

-    child.on('close', (code) => {
-      resolve({
-        ok: code === 0,
-        code: code || 0,
-        stdout,
-        stderr,
-        startedAt,
-        finishedAt: new Date(),
-        command: [command, ...args].join(' ')
+    child.on('close', (code, signal) => {
+      const finishedAt = new Date();
+      const exitCode = typeof code === 'number' ? code : 1;
+      let finalStderr = stderr;
+      if (signal) {
+        const signalMessage = `Process terminated by signal ${signal}`;
+        finalStderr = finalStderr ? `${finalStderr}\n${signalMessage}` : signalMessage;
+      }
+      resolve({
+        ok: exitCode === 0,
+        code: exitCode,
+        stdout,
+        stderr: finalStderr,
+        startedAt,
+        finishedAt,
+        command: [command, ...args].join(' '),
+        signal: signal || null

-          node -e "const fs=require('fs');const r=JSON.parse(fs.readFileSync('eval/report.json','utf8'));const score=Number((r.summary||{}).overallScore||0);if(score<40){console.error('[EvalBaselineGate] FAIL overallScore',score,'< 40');process.exit(1)};console.log('[EvalBaselineGate] PASS overallScore='+score);"
+          npm run eval:gate
+        env:
+          EVAL_GATE_MIN_SCORE: '40'

Conversation

GopiB9119 commented Mar 5, 2026

Summary

Validation

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 5, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants