Files
msd-core/tests/enh-2792-namespace-skills.test.cjs
Tom Boucher 5fdc950eb7 feat(#2792): namespace meta-skills + keyword-tag descriptions + context utilization guard (#2825)
* feat(#2792): namespace meta-skills retargeted at the post-#2790 surface

This branch is now based on #2790's HEAD (the consolidation PR) instead
of main, and every routing table targets the consolidated surface so a
user routed by a namespace meta-skill never lands at a deleted /
folded sub-skill.

Cross-PR inconsistencies the original PR #2825 carried (vs #2790):

  - ns-ideate routed to gsd-note / gsd-add-todo / gsd-add-backlog /
    gsd-plant-seed → all folded into gsd-capture by #2790. Now routes
    to gsd-capture (the parent picks the mode from the user's intent).
  - ns-context routed to gsd-scan and gsd-intel → folded into
    gsd-map-codebase --fast / --query by #2790. Now routes to those
    flag forms.
  - ns-manage routed all workspace intent to gsd-list-workspaces (a
    list-only entry) → CR also flagged the over-narrow target. #2790
    folds into gsd-workspace; routing now points there.
  - ns-workflow routed to gsd-research-phase → deleted outright by
    #2790. Removed.
  - ns-project routed to gsd-plan-milestone-gaps → deleted outright by
    #2790. Removed.
  - None of the namespaces previously surfaced #2790's new consolidated
    skills (gsd-capture, gsd-phase, gsd-config, gsd-workspace,
    gsd-progress). All five are now reachable through the routers.
  - extract_learnings → extract-learnings (canonicalized by #2858).

Defect fixes within the namespace skills:

  - Hyphen-form `name:` (gsd-workflow, …) per the canonical naming
    contract — the colon-form addressed CR's drift complaint.
  - `Skill` added to allowed-tools on every router. The body instructs
    "Invoke the matched skill directly using the Skill tool" — without
    Skill in the permission list the meta-skill cannot route at all.

New regression guard in tests/enh-2792-namespace-skills.test.cjs: every
gsd-* token in any namespace router's table column resolves to a
surviving commands/gsd/*.md file (or to a known consolidated parent for
flag-form targets like gsd-map-codebase --fast). This single test would
have caught every dead-end route the original PR shipped with.

Skill-count cap in tests/enh-2790-skill-consolidation.test.cjs now
filters out ns-*.md from its <= 63 cap. Namespace routers are
descriptor-only entries, not part of the consolidation surface that cap
is policing — they have their own contract in
tests/enh-2792-namespace-skills.test.cjs.

INVENTORY.md gains a "Namespace Meta-Skills" section with the 6 router
rows; INVENTORY-MANIFEST.json gains 6 entries; the headline count moves
59 → 65 to match.

Out of scope for this rebase: the gsd-health --context flag (PR #2825
advertised the contract but didn't implement it). That's a separate
feature concern and is left untouched here.

5908/5908 on `npm test`.

* feat(#2792): implement gsd-health --context utilization guard

The original PR #2825 advertised a `--context` flag on gsd-health with a
60%/70% utilization threshold table but never implemented the workflow
logic — CR caught it as a contract leak, the rebase deferred it. This
commit closes the gap with TDD red/green/refactor.

Math layer (pure):
  - get-shit-done/bin/lib/context-utilization.cjs
    classifyContextUtilization(tokensUsed, contextWindow) →
      { percent, state }
    State boundaries use the exact ratio:
      < 60% healthy / 60–70% warning / ≥ 70% critical (fracture point)
    Display percent rounded for humans. Throws TypeError on non-integer
    or out-of-range inputs.
  - STATES = Object.freeze({ HEALTHY, WARNING, CRITICAL }) exported
    so callers reference the names by symbol, not by literal string.

SDK CLI integration:
  - get-shit-done/bin/gsd-tools.cjs
    `validate context --tokens-used N --context-window M [--json]`
    routes to the classifier, owns the recommendation copy (the
    classifier intentionally does not — keeps the renderer free to
    evolve without touching the math layer or its tests), and uses
    core.output's rawValue path for the sync-flush guarantee.
  - sdk/src/query/validate.ts + sdk/src/query/index.ts
    TypeScript validateContext handler registered at 'validate.context'
    and 'validate context'. Mirrors the CJS classifier inline (15 lines
    of arithmetic; not worth a shared cross-language module).

User-facing wiring:
  - commands/gsd/health.md frontmatter advertises --context, body
    documents the three-state threshold table.
  - get-shit-done/workflows/health.md adds a `context_check` step
    that's reached only when --context is set. Step calls
    `gsd-sdk query validate.context` with self-reported tokensUsed and
    contextWindow, prints the SDK output verbatim, and ends. Includes
    a TEXT_MODE plain-text fallback for non-Claude runtimes per #2012.

Tests:
  - tests/context-utilization.test.cjs (17 tests) — pure-function
    contract: state thresholds at every boundary, percent rounding,
    input validation, return-shape (no recommendation field — that's
    the renderer's job).
  - tests/validate-context.test.cjs (9 tests) — SDK CLI plumbing:
    arg parsing errors, JSON vs human rendering, recommendation copy
    pinned per state.
  - tests/enh-2792-namespace-skills.test.cjs (4 new tests) — markdown
    contract: --context advertised in argument-hint, threshold table
    in command body, context_check step exists in workflow, step
    invokes gsd-sdk query validate.context with both flags.

Inventory bookkeeping:
  - docs/INVENTORY.md "CLI Modules" 31 → 32; new row for
    context-utilization.cjs.
  - docs/INVENTORY-MANIFEST.json mirror.

5939/5939 on `npm test`.
2026-04-30 01:04:41 -04:00

256 lines
11 KiB
JavaScript

'use strict';
// allow-test-rule: source-text-is-the-product
// commands/gsd/*.md files ARE what the runtime loads — testing their
// frontmatter content tests the deployed system-prompt contract.
const { describe, test } = require('node:test');
const assert = require('node:assert/strict');
const fs = require('node:fs');
const path = require('node:path');
const COMMANDS_DIR = path.join(__dirname, '..', 'commands', 'gsd');
const NAMESPACE_SKILLS = [
{ file: 'ns-workflow.md', name: 'gsd-workflow' },
{ file: 'ns-project.md', name: 'gsd-project' },
{ file: 'ns-review.md', name: 'gsd-review' },
{ file: 'ns-context.md', name: 'gsd-context' },
{ file: 'ns-manage.md', name: 'gsd-manage' },
{ file: 'ns-ideate.md', name: 'gsd-ideate' },
];
// Route targets named in any namespace body. The cross-reference test below
// asserts that every one of these resolves to a surviving command file or to
// a known consolidated parent (which absorbs flag-form invocations of folded
// skills, e.g. `gsd-map-codebase --fast` for the former `gsd-scan`).
const FLAG_FORM_PARENTS = new Set([
'gsd-code-review', // --fix absorbs former gsd-code-review-fix
'gsd-map-codebase', // --fast absorbs scan, --query absorbs intel
]);
/**
* Parse the leading YAML frontmatter block of a markdown file into a
* shallow `{ key: value }` map plus the trailing body. Splits on `\r?\n`
* for CRLF tolerance and uses trimmed-line equality for the `---`
* delimiters so whitespace-padded delimiter lines are accepted.
*/
function parseFrontmatter(content) {
const lines = content.split(/\r?\n/);
let openIdx = -1;
let closeIdx = -1;
for (let i = 0; i < lines.length; i += 1) {
if (lines[i].trim() === '---') {
if (openIdx === -1) openIdx = i;
else { closeIdx = i; break; }
}
}
assert.ok(openIdx !== -1 && closeIdx !== -1, 'frontmatter block must be delimited by --- on its own lines');
const fm = {};
for (const line of lines.slice(openIdx + 1, closeIdx)) {
const m = line.match(/^([A-Za-z][A-Za-z0-9_-]*):\s*(.*)$/);
if (!m) continue;
const [, key, raw] = m;
const value = raw.trim().replace(/^["']|["']$/g, '');
fm[key] = value;
}
fm._body = lines.slice(closeIdx + 1).join('\n');
return fm;
}
function readNamespaceFile(file) {
const filePath = path.join(COMMANDS_DIR, file);
assert.ok(fs.existsSync(filePath), `${file} must exist at ${filePath}`);
return { filePath, ...parseFrontmatter(fs.readFileSync(filePath, 'utf-8')) };
}
// ── Frontmatter contract ───────────────────────────────────────────────
describe('Namespace skill files exist with correct name', () => {
for (const { file, name } of NAMESPACE_SKILLS) {
test(`${file} — name field is hyphen-form ${name}`, () => {
const fm = readNamespaceFile(file);
assert.strictEqual(
fm.name,
name,
`name: in ${file} must be ${name} (hyphen form per #2858), got: ${fm.name}`,
);
});
}
});
describe('Namespace skill descriptions are keyword-tag format', () => {
for (const { file } of NAMESPACE_SKILLS) {
test(`${file} — description ≤ 60 chars`, () => {
const fm = readNamespaceFile(file);
assert.ok(
fm.description.length <= 60,
`${file} description must be ≤ 60 chars, got ${fm.description.length}: ${fm.description}`,
);
});
test(`${file} — description contains a pipe separator`, () => {
const fm = readNamespaceFile(file);
assert.ok(
fm.description.includes('|'),
`${file} description must contain | pipe separator, got: ${fm.description}`,
);
});
test(`${file} — description does not start with prose ("Use " / "This skill")`, () => {
const { description } = readNamespaceFile(file);
assert.ok(
!description.startsWith('Use ') && !description.startsWith('This skill'),
`${file} description must not start with "Use " or "This skill", got: ${description}`,
);
});
}
});
// ── allowed-tools must include Skill ──────────────────────────────────
describe('Namespace skills permit Skill execution', () => {
for (const { file } of NAMESPACE_SKILLS) {
test(`${file} — allowed-tools includes Skill`, () => {
const filePath = path.join(COMMANDS_DIR, file);
const raw = fs.readFileSync(filePath, 'utf-8');
const lines = raw.split(/\r?\n/);
const startIdx = lines.findIndex((l) => l.trim() === 'allowed-tools:');
assert.ok(startIdx !== -1, `${file} must declare an allowed-tools block`);
const tools = [];
for (let i = startIdx + 1; i < lines.length; i += 1) {
const m = lines[i].match(/^\s+-\s+(\S+)/);
if (!m) break;
tools.push(m[1]);
}
assert.ok(
tools.includes('Skill'),
`${file} body invokes the Skill tool but allowed-tools does not include Skill (got: ${tools.join(', ')})`,
);
});
}
});
// ── Body contains routing table ───────────────────────────────────────
describe('Namespace skill bodies carry a routing table', () => {
for (const { file } of NAMESPACE_SKILLS) {
test(`${file} — body contains "| User wants" table header`, () => {
const fm = readNamespaceFile(file);
const lines = fm._body.split('\n');
const hasHeader = lines.some((l) => l.includes('| User wants'));
assert.ok(hasHeader, `${file} body must contain a routing table starting with "| User wants"`);
});
test(`${file} — body has at least one Invoke target`, () => {
const fm = readNamespaceFile(file);
const hasInvoke = /\bgsd-[a-z-]+/i.test(fm._body);
assert.ok(hasInvoke, `${file} body must reference at least one gsd-* sub-skill`);
});
}
});
// ── Context guard contract on gsd-health ──────────────────────────────
// Asserts the `--context` surface promised by #2792 is wired through to
// both the command frontmatter and the workflow body. The classifier
// itself is covered by tests/context-utilization.test.cjs and the SDK
// CLI by tests/validate-context.test.cjs.
describe('gsd-health --context flag is wired into command + workflow', () => {
const HEALTH_CMD = path.join(COMMANDS_DIR, 'health.md');
const HEALTH_WORKFLOW = path.join(__dirname, '..', 'get-shit-done', 'workflows', 'health.md');
test('commands/gsd/health.md argument-hint advertises --context', () => {
const raw = fs.readFileSync(HEALTH_CMD, 'utf-8');
const fm = parseFrontmatter(raw);
assert.ok(
fm['argument-hint'] && fm['argument-hint'].includes('--context'),
`health.md argument-hint must include --context, got: ${fm['argument-hint']}`,
);
});
test('commands/gsd/health.md body documents the three-state utilization table', () => {
const raw = fs.readFileSync(HEALTH_CMD, 'utf-8');
const body = parseFrontmatter(raw)._body.toLowerCase();
assert.ok(body.includes('healthy'), 'body must name the healthy state');
assert.ok(body.includes('warning'), 'body must name the warning state');
assert.ok(body.includes('critical'), 'body must name the critical state');
assert.ok(
body.includes('60%') && body.includes('70%'),
'body must reference the 60% and 70% threshold boundaries',
);
});
test('get-shit-done/workflows/health.md has a context_check step', () => {
const raw = fs.readFileSync(HEALTH_WORKFLOW, 'utf-8');
assert.match(
raw,
/<step name="context_check">/,
'workflow must define a <step name="context_check"> branch',
);
});
test('workflow context_check invokes gsd-sdk query validate.context', () => {
const raw = fs.readFileSync(HEALTH_WORKFLOW, 'utf-8');
// Extract just the context_check step's body so a stray reference
// elsewhere in the file can't satisfy this assertion.
const stepMatch = raw.match(/<step name="context_check">([\s\S]*?)<\/step>/);
assert.ok(stepMatch, 'context_check step must be a closed <step>...</step> block');
const stepBody = stepMatch[1];
assert.match(
stepBody,
/gsd-sdk\s+query\s+validate\.context/,
'context_check must call `gsd-sdk query validate.context`',
);
assert.match(stepBody, /--tokens-used/, 'context_check must pass --tokens-used');
assert.match(stepBody, /--context-window/, 'context_check must pass --context-window');
});
});
// ── Cross-reference: every routed sub-skill must exist ─────────────────
// This is the regression guard the original PR lacked. Without it,
// post-#2790 consolidations can quietly invalidate router targets again.
describe('Namespace router targets resolve to surviving skills', () => {
// Build the post-consolidation surviving set once.
const surviving = new Set();
for (const f of fs.readdirSync(COMMANDS_DIR)) {
if (!f.endsWith('.md')) continue;
const base = f.replace(/\.md$/, '');
if (base.startsWith('ns-')) continue; // namespace routers themselves
surviving.add(`gsd-${base}`);
// The PR #2858 rename canonicalized extract_learnings → extract-learnings.
// Until #2790 rebases onto current main, accept either source filename
// as resolving to the canonical hyphenated identifier.
if (base === 'extract_learnings') surviving.add('gsd-extract-learnings');
}
for (const { file } of NAMESPACE_SKILLS) {
test(`${file} — every routing target resolves`, () => {
const fm = readNamespaceFile(file);
// Extract every gsd-<name> token that appears in a table-row right column.
// Strip flag suffixes (`gsd-foo --bar` → `gsd-foo`) before resolving.
const targets = new Set();
for (const line of fm._body.split('\n')) {
// Only consider markdown table data rows: lines that start with `|`
// and have content between pipes. Skip header / separator rows.
if (!line.startsWith('|') || /^\|[\s\-:|]+\|?\s*$/.test(line)) continue;
const cells = line.split('|').map((c) => c.trim()).filter(Boolean);
if (cells.length < 2) continue;
for (const m of cells[cells.length - 1].matchAll(/\bgsd-[a-z][a-z0-9-]*/g)) {
targets.add(m[0]);
}
}
assert.ok(targets.size > 0, `${file} routing table must reference at least one gsd-* target`);
const unresolved = [...targets].filter(
(t) => !surviving.has(t) && !FLAG_FORM_PARENTS.has(t),
);
assert.deepStrictEqual(
unresolved,
[],
`${file} routes to skills that don't exist in commands/gsd/: ${unresolved.join(', ')}`,
);
});
}
});