Harden verify-work checkpoint rendering

This commit is contained in:
3metaJun
2026-03-24 01:16:27 +08:00
committed by Tom Boucher
parent f43e0237f2
commit aaaa8e96fe
7 changed files with 249 additions and 15 deletions

View File

@@ -67,6 +67,7 @@
*
* UAT Audit:
* audit-uat Scan all phases for unresolved UAT/verification items
* uat render-checkpoint --file <path> Render the current UAT checkpoint block
*
* Scaffolding:
* scaffold context --phase <N> Create CONTEXT.md template
@@ -660,6 +661,18 @@ async function runCommand(command, args, cwd, raw) {
break;
}
case 'uat': {
const subcommand = args[1];
const uat = require('./lib/uat.cjs');
if (subcommand === 'render-checkpoint') {
const options = parseNamedArgs(args, ['file']);
uat.cmdRenderCheckpoint(cwd, options, raw);
} else {
error('Unknown uat subcommand. Available: render-checkpoint');
}
break;
}
case 'stats': {
const subcommand = args[1] || 'json';
commands.cmdStats(cwd, subcommand, raw);

View File

@@ -223,6 +223,32 @@ function sanitizeForPrompt(text) {
return sanitized;
}
/**
* Sanitize text that will be displayed back to the user.
* Removes protocol-like leak markers that should never surface in checkpoints.
*
* @param {string} text - Text to sanitize
* @returns {string} Sanitized text
*/
function sanitizeForDisplay(text) {
if (!text || typeof text !== 'string') return text;
let sanitized = sanitizeForPrompt(text);
const protocolLeakPatterns = [
/^\s*(?:assistant|user|system)\s+to=[^:\s]+:[^\n]+$/i,
/^\s*(?:assistant|user|system)\s+to=all:[^\n]+$/i,
/^\s*<\|(?:assistant|user|system)[^|]*\|>\s*$/i,
];
sanitized = sanitized
.split('\n')
.filter(line => !protocolLeakPatterns.some(pattern => pattern.test(line)))
.join('\n');
return sanitized;
}
// ─── Shell Safety ───────────────────────────────────────────────────────────
/**
@@ -343,6 +369,7 @@ module.exports = {
INJECTION_PATTERNS,
scanForInjection,
sanitizeForPrompt,
sanitizeForDisplay,
// Shell safety
validateShellArg,

View File

@@ -9,6 +9,7 @@ const fs = require('fs');
const path = require('path');
const { output, error, getMilestonePhaseFilter, planningDir, toPosixPath } = require('./core.cjs');
const { extractFrontmatter } = require('./frontmatter.cjs');
const { requireSafePath, sanitizeForDisplay } = require('./security.cjs');
function cmdAuditUat(cwd, raw) {
const phasesDir = path.join(planningDir(cwd), 'phases');
@@ -90,6 +91,92 @@ function cmdAuditUat(cwd, raw) {
output({ results, summary }, raw);
}
function cmdRenderCheckpoint(cwd, options = {}, raw) {
const filePath = options.file;
if (!filePath) {
error('UAT file required: use uat render-checkpoint --file <path>');
}
const resolvedPath = requireSafePath(filePath, cwd, 'UAT file', { allowAbsolute: true });
if (!fs.existsSync(resolvedPath)) {
error(`UAT file not found: ${filePath}`);
}
const content = fs.readFileSync(resolvedPath, 'utf-8');
const currentTest = parseCurrentTest(content);
if (currentTest.complete) {
error('UAT session is already complete; no pending checkpoint to render');
}
const checkpoint = buildCheckpoint(currentTest);
output({
file_path: toPosixPath(path.relative(cwd, resolvedPath)),
test_number: currentTest.number,
test_name: currentTest.name,
checkpoint,
}, raw, checkpoint);
}
function parseCurrentTest(content) {
const currentTestMatch = content.match(/##\s*Current Test\s*(?:\n<!--[\s\S]*?-->)?\n([\s\S]*?)(?=\n##\s|$)/i);
if (!currentTestMatch) {
error('UAT file is missing a Current Test section');
}
const section = currentTestMatch[1].trimEnd();
if (!section.trim()) {
error('Current Test section is empty');
}
if (/\[testing complete\]/i.test(section)) {
return { complete: true };
}
const numberMatch = section.match(/^number:\s*(\d+)\s*$/m);
const nameMatch = section.match(/^name:\s*(.+)\s*$/m);
const expectedBlockMatch = section.match(/^expected:\s*\|\n([\s\S]*?)(?=^\w[\w-]*:\s|\Z)/m);
const expectedInlineMatch = section.match(/^expected:\s*(.+)\s*$/m);
if (!numberMatch || !nameMatch || (!expectedBlockMatch && !expectedInlineMatch)) {
error('Current Test section is malformed');
}
let expected;
if (expectedBlockMatch) {
expected = expectedBlockMatch[1]
.split('\n')
.map(line => line.replace(/^ {2}/, ''))
.join('\n')
.trim();
} else {
expected = expectedInlineMatch[1].trim();
}
return {
complete: false,
number: parseInt(numberMatch[1], 10),
name: sanitizeForDisplay(nameMatch[1].trim()),
expected: sanitizeForDisplay(expected),
};
}
function buildCheckpoint(currentTest) {
return [
'╔══════════════════════════════════════════════════════════════╗',
'║ CHECKPOINT: Verification Required ║',
'╚══════════════════════════════════════════════════════════════╝',
'',
`**Test ${currentTest.number}: ${currentTest.name}**`,
'',
currentTest.expected,
'',
'──────────────────────────────────────────────────────────────',
'Type `pass` or describe what\'s wrong.',
'──────────────────────────────────────────────────────────────',
].join('\n');
}
function parseUatItems(content) {
const items = [];
// Match test blocks: ### N. Name\nexpected: ...\nresult: ...\n
@@ -186,4 +273,9 @@ function categorizeItem(result, reason, blockedBy) {
return 'unknown';
}
module.exports = { cmdAuditUat };
module.exports = {
cmdAuditUat,
cmdRenderCheckpoint,
parseCurrentTest,
buildCheckpoint,
};