feat(catalog): T4 — catalog trim + proactive-suggestions.json (Phase A.4)

Shortens frontmatter `description:` in every Claude SKILL.md to a single
lead sentence + (gstack) tag. The routing prose ("Use when asked to...",
"Proactively suggest...") and voice triggers move to a "## When to invoke"
body section so they remain discoverable inside the skill. A per-run
registry at scripts/proactive-suggestions.json aggregates the routing/
voice text for all 52 skills so agents can pull guidance on demand
without paying for it in the always-loaded catalog.

Build flag --catalog-mode=full restores v1.44 legacy behavior (full
multi-line descriptions in frontmatter). Default is trim.

splitCatalogDescription() extracts: lead sentence, routing paragraphs,
voice-triggers line, (gstack) tag presence. Short descriptions (<120
chars, already trimmed) are skipped via a guard so re-runs are idempotent.

Measured impact (vs v1.44.1 baseline):
- Catalog tokens (sum of description bytes / 4): 9,319 → 4,045  (-56.6%)
- Total SKILL.md corpus bytes:                   2,915 KB → 2,880 KB (-1.2%)
- Routing prose preserved as in-skill "## When to invoke" sections
- 52 skill entries in scripts/proactive-suggestions.json (on-demand registry)

The corpus drop is small because catalog trim MOVES text from frontmatter
to body, it doesn't delete it. The headline win is the catalog: the
always-loaded system prompt surface drops by more than half.

Test plan:
- bun test test/gen-skill-docs.test.ts: 389 pass, 0 fail
- Manual: ship/SKILL.md frontmatter description is now ONE line ending
  with `(gstack)`; allowed-tools field on next line (YAML well-formed)
- Manual: scripts/proactive-suggestions.json contains 52 entries
- bun run gen:skill-docs --catalog-mode=full restores legacy behavior

53 files changed (52 SKILL.md across hosts + the new proactive-suggestions.json).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Garry Tan
2026-05-25 20:35:15 -07:00
parent da5f26872f
commit 0d68ef1a39
54 changed files with 1078 additions and 398 deletions
+215 -3
View File
@@ -59,6 +59,23 @@ const MODEL_ARG_VAL: Model = (() => {
return resolved;
})();
// ─── Catalog Mode (v1.45.0.0 T4) ────────────────────────────
// 'trim' (default): shorten frontmatter description to lead sentence,
// move routing/voice prose into a "## When to invoke" body section, and
// emit scripts/proactive-suggestions.json (single file across all skills).
// 'full': legacy v1.44 behavior — full description stays in frontmatter.
const CATALOG_MODE_ARG = process.argv.find(a => a.startsWith('--catalog-mode'));
const CATALOG_MODE: 'trim' | 'full' = (() => {
if (!CATALOG_MODE_ARG) return 'trim';
const val = CATALOG_MODE_ARG.includes('=')
? CATALOG_MODE_ARG.split('=')[1]
: process.argv[process.argv.indexOf(CATALOG_MODE_ARG) + 1];
if (val !== 'trim' && val !== 'full') {
throw new Error(`Unknown catalog mode: ${val}. Use 'trim' (default) or 'full'.`);
}
return val;
})();
// ─── Explain-level Overlay ──────────────────────────────────
// --explain-level=terse compresses preamble prose (writing-style, completeness,
// confusion-protocol, context-health) to a single pointer line at gen time.
@@ -190,6 +207,160 @@ function processVoiceTriggers(content: string): string {
// Export for testing
export { extractVoiceTriggers, processVoiceTriggers };
// ─── Catalog Trim (v1.45.0.0 T4) ─────────────────────────────
//
// Frontmatter `description:` blocks today pack: a one-line outcome, "Use when
// asked to..." voice triggers, "Proactively..." routing guidance, and a
// "(gstack)" tag. This pile is the always-loaded catalog surface — every
// session pays for the full text. The catalog trim splits the description
// into a one-line catalog entry (lead sentence + "(gstack)") that stays in
// the frontmatter, and a "## When to invoke" body section that holds the
// routing/voice triggers prose for in-skill discovery. A registry written
// to scripts/proactive-suggestions.json (one entry per skill) makes routing
// available to agents that need it without paying the always-loaded cost.
//
// Opt-out: `--catalog-mode=full` keeps v1.44 behavior (no trim, full
// description in frontmatter). Use when debugging routing regressions or
// when shipping skills to hosts that depend on the legacy fat catalog.
export interface CatalogParts {
lead: string; // First sentence — kept in catalog
routingProse: string; // "Use when asked to...", "Proactively..." paragraphs
voiceLine: string | null; // "Voice triggers (speech-to-text aliases): ..." line if present
hasGstackTag: boolean;
}
export function splitCatalogDescription(description: string): CatalogParts {
// Voice triggers line (folded in by processVoiceTriggers earlier)
const voiceMatch = description.match(/Voice triggers \(speech-to-text aliases\):[^\n]+/);
const voiceLine = voiceMatch ? voiceMatch[0] : null;
let working = voiceLine ? description.replace(voiceLine, '').trim() : description.trim();
const hasGstackTag = /\(gstack\)/.test(working);
if (hasGstackTag) working = working.replace(/\(gstack\)/, '').trim();
// Lead = first sentence (up to first period followed by space or end of string).
// We tolerate sentences with embedded periods (URLs, "v1.45.0.0") by requiring
// the period to be followed by whitespace OR end-of-text.
// First normalize to single-line for sentence detection, then back out.
const collapsed = working.replace(/\s+/g, ' ').trim();
const sentenceMatch = collapsed.match(/^([^.!?]*[.!?])(?:\s|$)/);
let lead = sentenceMatch ? sentenceMatch[1].trim() : collapsed.split(/\s/).slice(0, 20).join(' ');
// If the lead would be too long, trim to the first 140 chars at a word boundary
if (lead.length > 200) {
const trunc = lead.slice(0, 197);
const lastSpace = trunc.lastIndexOf(' ');
lead = (lastSpace > 60 ? trunc.slice(0, lastSpace) : trunc) + '...';
}
const leadInCollapsed = collapsed.indexOf(lead);
const routingCollapsed = leadInCollapsed >= 0
? collapsed.slice(leadInCollapsed + lead.length).trim()
: '';
// Restore line breaks for routing prose by mapping back to original layout.
// Use original whitespace structure where possible; fall back to collapsed.
let routingProse = routingCollapsed;
// Try to recover the multi-line layout: split working at the lead boundary.
const collapsedLeadIdx = working.replace(/\s+/g, ' ').indexOf(lead);
if (collapsedLeadIdx >= 0) {
// Walk the original working string until we've consumed lead.length collapsed chars
let consumed = 0;
let cut = 0;
for (let i = 0; i < working.length && consumed < collapsedLeadIdx + lead.length; i++) {
if (/\s/.test(working[i])) {
if (i === 0 || /\s/.test(working[i - 1])) continue;
consumed += 1;
} else {
consumed += 1;
}
cut = i + 1;
}
const tail = working.slice(cut).trim();
if (tail.length > 0) routingProse = tail;
}
return { lead, routingProse, voiceLine, hasGstackTag };
}
/** Build the catalog-trimmed `description:` block. */
export function buildTrimmedDescription(parts: CatalogParts): string {
const lead = parts.lead.trim();
const suffix = parts.hasGstackTag ? ' (gstack)' : '';
return `${lead}${suffix}`;
}
/** Build the body section that holds the routing/voice prose. */
export function buildWhenToInvokeSection(parts: CatalogParts): string {
const lines: string[] = ['## When to invoke this skill', ''];
if (parts.routingProse) {
lines.push(parts.routingProse);
lines.push('');
}
if (parts.voiceLine) {
lines.push(parts.voiceLine);
lines.push('');
}
return lines.join('\n');
}
/**
* Apply catalog trim to a SKILL.md body:
* - shorten frontmatter `description:` to lead + (gstack)
* - insert "## When to invoke" body section AFTER the generated header
* (so it lands near the top of body content, where routing guidance
* belongs)
*
* Returns the rewritten content plus the parts (used for proactive-suggestions
* JSON aggregation at the end of the run).
*/
export function applyCatalogTrim(content: string, skillName: string): { content: string; parts: CatalogParts } | null {
// Locate description block in frontmatter
if (!content.startsWith('---\n')) return null;
const fmEnd = content.indexOf('\n---', 4);
if (fmEnd === -1) return null;
const frontmatter = content.slice(4, fmEnd);
// Match `description: |` block + indented body lines
const descMatch = frontmatter.match(/^description:\s*\|?\s*\n((?:\s{2,}.*(?:\n|$))+)/m)
|| frontmatter.match(/^description:\s+(.+)$/m);
if (!descMatch) return null;
// Extract full description text
let descText: string;
if (descMatch[0].startsWith('description: |') || /^description:\s*\|/.test(descMatch[0])) {
descText = descMatch[1].split('\n').map(l => l.replace(/^\s{2}/, '')).join('\n').trim();
} else {
descText = descMatch[1].trim();
}
// Skip skills with very short descriptions (already trimmed or no routing prose).
// Below ~120 chars, splitting adds no value.
if (descText.length < 120) return null;
const parts = splitCatalogDescription(descText);
// If lead + (gstack) is already most of the text, no trim needed.
const trimmedLen = buildTrimmedDescription(parts).length;
if (trimmedLen >= descText.length - 20) return null;
// Replace description in frontmatter — keep trailing newline so the next
// YAML field doesn't collide on the same line as the description value.
const newDesc = buildTrimmedDescription(parts);
const newFrontmatter = frontmatter.replace(descMatch[0], `description: ${newDesc}\n`);
let newContent = '---\n' + newFrontmatter + content.slice(fmEnd);
// Insert body section after frontmatter (after the closing ---\n and any
// existing GENERATED header). We insert before the first non-comment line.
const bodyStart = newContent.indexOf('\n---\n') + 5;
const whenToInvoke = '\n' + buildWhenToInvokeSection(parts).trim() + '\n';
// Skip past the generated header if present (it lives after frontmatter close)
const headerMatch = newContent.slice(bodyStart).match(/^(<!--[^>]*-->\s*\n)+/);
const insertAt = bodyStart + (headerMatch ? headerMatch[0].length : 0);
newContent = newContent.slice(0, insertAt) + whenToInvoke + '\n' + newContent.slice(insertAt);
return { content: newContent, parts };
}
const OPENAI_SHORT_DESCRIPTION_LIMIT = 120;
function condenseOpenAIShortDescription(description: string): string {
@@ -419,7 +590,7 @@ function processExternalHost(
return { content: result, outputPath, outputDir, symlinkLoop };
}
function processTemplate(tmplPath: string, host: Host = 'claude'): { outputPath: string; content: string; symlinkLoop?: boolean } {
function processTemplate(tmplPath: string, host: Host = 'claude'): { outputPath: string; content: string; symlinkLoop?: boolean; catalogParts?: CatalogParts | null } {
const tmplContent = fs.readFileSync(tmplPath, 'utf-8');
const relTmplPath = path.relative(ROOT, tmplPath);
let outputPath = tmplPath.replace(/\.tmpl$/, '');
@@ -503,7 +674,17 @@ function processTemplate(tmplPath: string, host: Host = 'claude'): { outputPath:
content = header + content;
}
return { outputPath, content, symlinkLoop };
// Catalog trim (Claude only — external hosts have their own frontmatter shapes)
let catalogParts: CatalogParts | null = null;
if (host === 'claude' && CATALOG_MODE === 'trim') {
const trimmed = applyCatalogTrim(content, skillName);
if (trimmed) {
content = trimmed.content;
catalogParts = trimmed.parts;
}
}
return { outputPath, content, symlinkLoop, catalogParts };
}
// ─── Main ───────────────────────────────────────────────────
@@ -523,6 +704,14 @@ for (const currentHost of hostsToRun) {
let hasChanges = false;
const tokenBudget: Array<{ skill: string; lines: number; tokens: number }> = [];
// T4 catalog trim: collect routing/voice parts across all Claude skills,
// then write scripts/proactive-suggestions.json once per gen-skill-docs run.
const proactiveAggregate: Record<string, {
lead: string;
routing: string;
voice_line: string | null;
}> = {};
const currentHostConfig = getHostConfig(currentHost);
for (const tmplPath of findTemplates()) {
const dir = path.basename(path.dirname(tmplPath));
@@ -536,7 +725,15 @@ for (const currentHost of hostsToRun) {
if (currentHostConfig.generation.skipSkills.includes(dir)) continue;
}
const { outputPath, content, symlinkLoop } = processTemplate(tmplPath, currentHost);
const { outputPath, content, symlinkLoop, catalogParts } = processTemplate(tmplPath, currentHost);
if (catalogParts) {
const key = dir === '' ? 'gstack' : dir;
proactiveAggregate[key] = {
lead: catalogParts.lead,
routing: catalogParts.routingProse,
voice_line: catalogParts.voiceLine,
};
}
const relOutput = path.relative(ROOT, outputPath);
if (symlinkLoop) {
@@ -640,6 +837,21 @@ The orchestrator will persist the plan link to its own memory/knowledge store.
failures.push({ host: currentHost, error: new Error('Stale files detected') });
}
// T4 catalog trim: write aggregated proactive-suggestions.json (Claude only).
// The JSON registry lets agents pull voice triggers / routing prose for any
// skill on demand instead of paying for it always-loaded in the catalog.
if (currentHost === 'claude' && CATALOG_MODE === 'trim' && Object.keys(proactiveAggregate).length > 0 && !DRY_RUN) {
const proactivePath = path.join(ROOT, 'scripts', 'proactive-suggestions.json');
const payload = {
$schema: 'https://gstack.dev/schemas/proactive-suggestions.json',
generated_at: new Date().toISOString(),
catalog_mode: 'trim',
note: 'Routing / voice-trigger prose extracted from SKILL.md frontmatter descriptions during catalog trim. Loaded on demand when routing guidance is needed.',
skills: proactiveAggregate,
};
fs.writeFileSync(proactivePath, JSON.stringify(payload, null, 2) + '\n');
}
// Print token budget summary
if (!DRY_RUN && tokenBudget.length > 0) {
tokenBudget.sort((a, b) => b.lines - a.lines);