Flag any single active agent whose description exceeds a 500-char soft cap (the same bloat threshold TOK pattern F applies to SKILL.md descriptions). Every char of an agent description re-enters context in the always-loaded agent listing on every turn, so a long one is a per-turn cost. Framing is deliberately ADVISORY, severity low: unlike CA-SKL-001 agents have NO verified per-description cap, so nothing is dropped — this is a bloat advisory, never a truncation claim. Per-agent advisories are emitted BEFORE the aggregate roll-up (mirrors SKL 001->002). - PER_AGENT_DESC_SOFT_CAP (=500) added to lib/agent-listing-budget.mjs as the single source of truth, disclosed as a heuristic. - Per-agent loop reuses measureActiveAgentListing()'s agents[] (name, source, pluginName, path, descLength) — no new enumeration. - 5 new tests (fire/strict-boundary/no-truncation-claim/recommendation/order). Byte-stable: AGT already stripped from frozen v5.0.0 baselines; SC-5 unchanged (HOME-scoped, hermetic HOME has no agents). Suite green: 1181 tests. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
283 lines
13 KiB
JavaScript
283 lines
13 KiB
JavaScript
import { describe, it } from 'node:test';
|
|
import assert from 'node:assert/strict';
|
|
import { join } from 'node:path';
|
|
import { mkdir, writeFile, rm } from 'node:fs/promises';
|
|
import { tmpdir } from 'node:os';
|
|
import { resetCounter } from '../../scanners/lib/output.mjs';
|
|
import { scan } from '../../scanners/agent-listing-scanner.mjs';
|
|
import { PER_AGENT_DESC_SOFT_CAP } from '../../scanners/lib/agent-listing-budget.mjs';
|
|
|
|
/**
|
|
* AGT scanner — agent-listing always-loaded token budget.
|
|
*
|
|
* Claude Code injects a listing of every active agent's name+description into the
|
|
* system prompt so the model knows which subagents it can delegate to. With ~100
|
|
* agents that listing is a large always-loaded cost. UNLIKE the skill listing,
|
|
* the agent-listing mechanism is NOT documented (agents are absent from CC's
|
|
* deferred / always-loaded context breakdown), so AGT findings are an INFERRED
|
|
* upper-bound estimate, not a verified truncation like SKL-001. The aggregate
|
|
* budget is a config-audit heuristic anchored on a conservative 200k window.
|
|
*
|
|
* Token heuristic mirrors estimateTokens('markdown'): ceil(chars/4).
|
|
* 16 agents * 1000 desc-chars = 16000 chars -> 4000 tok == budget (NOT over).
|
|
* 17 agents * 1000 desc-chars = 17000 chars -> 4250 tok > budget (over).
|
|
*/
|
|
const AGGREGATE_BUDGET_TOKENS = 4000; // heuristic: 0.02 * 200_000 (anchored, disclosed)
|
|
|
|
function uniqueDir(suffix) {
|
|
return join(tmpdir(), `config-audit-agt-${suffix}-${Date.now()}-${Math.random().toString(36).slice(2, 8)}`);
|
|
}
|
|
|
|
/**
|
|
* The AGT scanner enumerates active agents via process.env.HOME. Tests must
|
|
* override HOME, run, restore — never rely on the developer's real ~/.claude.
|
|
*/
|
|
async function runScannerWithHome(home) {
|
|
resetCounter();
|
|
const original = process.env.HOME;
|
|
process.env.HOME = home;
|
|
try {
|
|
return await scan('/unused', { files: [] });
|
|
} finally {
|
|
process.env.HOME = original;
|
|
}
|
|
}
|
|
|
|
/** Write one user agent (name + description of `descLen` chars) into a HOME. */
|
|
async function addUserAgent(home, name, descLen) {
|
|
const dir = join(home, '.claude', 'agents');
|
|
await mkdir(dir, { recursive: true });
|
|
const desc = 'a'.repeat(descLen);
|
|
await writeFile(
|
|
join(dir, `${name}.md`),
|
|
`---\nname: ${name}\ndescription: ${desc}\n---\nYou are the ${name} agent. Body of the agent prompt.\n`,
|
|
);
|
|
}
|
|
|
|
/** Build a fake HOME with `count` user agents, each description `descLen` chars. */
|
|
async function homeWithNUserAgents(count, descLen, prefix = 'agg') {
|
|
const home = uniqueDir(`${prefix}-${count}x${descLen}`);
|
|
for (let i = 0; i < count; i++) {
|
|
await addUserAgent(home, `${prefix}${i}`, descLen);
|
|
}
|
|
return home;
|
|
}
|
|
|
|
/** Build a fake HOME with a single user agent whose description is `descLen` chars. */
|
|
async function homeWithOneAgent(descLen, name = 'solo') {
|
|
const home = uniqueDir(`one-${name}-${descLen}`);
|
|
await addUserAgent(home, name, descLen);
|
|
return home;
|
|
}
|
|
|
|
const findAggregate = (findings) => findings.find(f => /agent listing/i.test(f.title));
|
|
// Per-agent advisories are every finding that is NOT the aggregate roll-up.
|
|
const findPerAgent = (findings) => findings.filter(f => !/agent listing/i.test(f.title));
|
|
|
|
describe('AGT scanner — basic structure', () => {
|
|
it('reports scanner prefix AGT', async () => {
|
|
const home = uniqueDir('empty');
|
|
try {
|
|
await mkdir(join(home, '.claude'), { recursive: true });
|
|
const result = await runScannerWithHome(home);
|
|
assert.equal(result.scanner, 'AGT');
|
|
} finally {
|
|
await rm(home, { recursive: true, force: true });
|
|
}
|
|
});
|
|
|
|
it('finding IDs match CA-AGT-NNN pattern', async () => {
|
|
const home = await homeWithNUserAgents(17, 1000);
|
|
try {
|
|
const result = await runScannerWithHome(home);
|
|
for (const f of result.findings) assert.match(f.id, /^CA-AGT-\d{3}$/);
|
|
} finally {
|
|
await rm(home, { recursive: true, force: true });
|
|
}
|
|
});
|
|
});
|
|
|
|
describe('AGT scanner — aggregate always-loaded budget', () => {
|
|
it('fires a low-severity finding when the agent listing exceeds the 200k-anchored budget', async () => {
|
|
const home = await homeWithNUserAgents(17, 1000); // 17000 chars -> ~4250 tok > 4000
|
|
try {
|
|
const result = await runScannerWithHome(home);
|
|
const agg = findAggregate(result.findings);
|
|
assert.ok(agg, `expected an aggregate finding; got: ${result.findings.map(f => f.title).join(' | ')}`);
|
|
assert.equal(agg.severity, 'low', `aggregate should be low severity, got ${agg.severity}`);
|
|
assert.match(agg.id, /^CA-AGT-\d{3}$/);
|
|
} finally {
|
|
await rm(home, { recursive: true, force: true });
|
|
}
|
|
});
|
|
|
|
it('does NOT fire when only a few small agents are active', async () => {
|
|
const home = await homeWithNUserAgents(3, 1000); // 3000 chars -> 750 tok < 4000
|
|
try {
|
|
const result = await runScannerWithHome(home);
|
|
assert.equal(findAggregate(result.findings), undefined,
|
|
`expected no aggregate finding; got: ${result.findings.map(f => f.title).join(' | ')}`);
|
|
} finally {
|
|
await rm(home, { recursive: true, force: true });
|
|
}
|
|
});
|
|
|
|
it('treats the budget as a strict boundary: 16x1000 (at budget) no fire, 17x1000 (over) fires', async () => {
|
|
const atBudget = await homeWithNUserAgents(16, 1000, 'at'); // 16000 chars -> 4000 tok == budget
|
|
try {
|
|
const result = await runScannerWithHome(atBudget);
|
|
assert.equal(findAggregate(result.findings), undefined,
|
|
'aggregate exactly at the budget must not fire (strictly-greater rule)');
|
|
} finally {
|
|
await rm(atBudget, { recursive: true, force: true });
|
|
}
|
|
const overBudget = await homeWithNUserAgents(17, 1000, 'over'); // 17000 chars -> 4250 tok
|
|
try {
|
|
const result = await runScannerWithHome(overBudget);
|
|
assert.ok(findAggregate(result.findings), 'aggregate one step over the budget must fire');
|
|
} finally {
|
|
await rm(overBudget, { recursive: true, force: true });
|
|
}
|
|
});
|
|
|
|
it('DISCLOSES the inferred mechanism + upper-bound estimate + anchor (intellectual honesty)', async () => {
|
|
const home = await homeWithNUserAgents(17, 1000);
|
|
try {
|
|
const result = await runScannerWithHome(home);
|
|
const agg = findAggregate(result.findings);
|
|
assert.ok(agg, 'expected an aggregate finding');
|
|
const ev = String(agg.evidence);
|
|
assert.match(ev, /infer/i, 'evidence must disclose the always-loaded mechanism is INFERRED, not documented');
|
|
assert.match(ev, /estimate|upper.?bound/i, 'evidence must flag the figure as an estimate / upper bound');
|
|
assert.match(ev, new RegExp(String(AGGREGATE_BUDGET_TOKENS)), 'evidence must state the budget');
|
|
assert.match(ev, /200k/, 'evidence must anchor the budget on a 200k window');
|
|
assert.match(ev, /1,000,000|1M|20,000/, 'evidence must note the 1M-context scaling');
|
|
} finally {
|
|
await rm(home, { recursive: true, force: true });
|
|
}
|
|
});
|
|
|
|
it('recommends the reduction levers (disable unused plugins / trim descriptions)', async () => {
|
|
const home = await homeWithNUserAgents(17, 1000);
|
|
try {
|
|
const result = await runScannerWithHome(home);
|
|
const agg = findAggregate(result.findings);
|
|
assert.ok(agg, 'expected an aggregate finding to carry remediation');
|
|
const rec = String(agg.recommendation);
|
|
assert.match(rec, /disable|plugin/i, 'recommendation should mention disabling unused plugins');
|
|
assert.match(rec, /trim|description/i, 'recommendation should mention trimming descriptions');
|
|
} finally {
|
|
await rm(home, { recursive: true, force: true });
|
|
}
|
|
});
|
|
|
|
it('an empty HOME (no agents) yields zero findings', async () => {
|
|
const home = uniqueDir('noagents');
|
|
try {
|
|
await mkdir(join(home, '.claude'), { recursive: true });
|
|
const result = await runScannerWithHome(home);
|
|
assert.equal(result.findings.length, 0);
|
|
} finally {
|
|
await rm(home, { recursive: true, force: true });
|
|
}
|
|
});
|
|
});
|
|
|
|
/**
|
|
* Per-agent description advisory (the B1-rest companion to the aggregate).
|
|
*
|
|
* A single agent with a long description is flagged on its own: every char of
|
|
* that description re-enters context in the always-loaded agent listing on every
|
|
* turn. UNLIKE CA-SKL-001, this is NOT a truncation finding — agents have no
|
|
* verified per-description cap — so the advisory is a soft heuristic (low, the
|
|
* same 500-char bloat threshold TOK pattern F uses for SKILL.md descriptions),
|
|
* not a hard "Claude Code drops the tail" claim.
|
|
*/
|
|
describe('AGT scanner — per-agent description advisory', () => {
|
|
it('fires a low-severity advisory when one agent description exceeds the soft cap', async () => {
|
|
// One agent, description over the cap; aggregate stays well under budget so
|
|
// the per-agent advisory is the only finding (clean isolation).
|
|
const home = await homeWithOneAgent(PER_AGENT_DESC_SOFT_CAP + 100);
|
|
try {
|
|
const result = await runScannerWithHome(home);
|
|
assert.equal(findAggregate(result.findings), undefined,
|
|
'a single small-aggregate agent must not trip the aggregate budget');
|
|
const perAgent = findPerAgent(result.findings);
|
|
assert.equal(perAgent.length, 1,
|
|
`expected exactly one per-agent advisory; got: ${result.findings.map(f => f.title).join(' | ')}`);
|
|
assert.equal(perAgent[0].severity, 'low', `per-agent advisory should be low, got ${perAgent[0].severity}`);
|
|
assert.match(perAgent[0].id, /^CA-AGT-\d{3}$/);
|
|
assert.match(`${perAgent[0].description} ${perAgent[0].evidence}`, /solo/,
|
|
'the advisory should name the offending agent');
|
|
} finally {
|
|
await rm(home, { recursive: true, force: true });
|
|
}
|
|
});
|
|
|
|
it('treats the soft cap as a strict boundary: at-cap no fire, one-over fires', async () => {
|
|
const atCap = await homeWithOneAgent(PER_AGENT_DESC_SOFT_CAP, 'atcap');
|
|
try {
|
|
const result = await runScannerWithHome(atCap);
|
|
assert.equal(findPerAgent(result.findings).length, 0,
|
|
'a description exactly at the soft cap must not fire (strictly-greater rule)');
|
|
} finally {
|
|
await rm(atCap, { recursive: true, force: true });
|
|
}
|
|
const overCap = await homeWithOneAgent(PER_AGENT_DESC_SOFT_CAP + 1, 'overcap');
|
|
try {
|
|
const result = await runScannerWithHome(overCap);
|
|
assert.equal(findPerAgent(result.findings).length, 1,
|
|
'a description one char over the soft cap must fire');
|
|
} finally {
|
|
await rm(overCap, { recursive: true, force: true });
|
|
}
|
|
});
|
|
|
|
it('does NOT claim truncation and DISCLOSES the advisory/heuristic framing (intellectual honesty)', async () => {
|
|
const home = await homeWithOneAgent(PER_AGENT_DESC_SOFT_CAP + 100);
|
|
try {
|
|
const result = await runScannerWithHome(home);
|
|
const advisory = findPerAgent(result.findings)[0];
|
|
assert.ok(advisory, 'expected a per-agent advisory');
|
|
const text = `${advisory.description} ${advisory.evidence}`;
|
|
assert.doesNotMatch(text, /truncat/i,
|
|
'agents have no verified cap — the advisory must NOT claim Claude Code truncates the description');
|
|
assert.match(text, /every turn|always.?loaded/i,
|
|
'the advisory must explain the description re-enters context every turn');
|
|
assert.match(String(advisory.evidence), new RegExp(String(PER_AGENT_DESC_SOFT_CAP)),
|
|
'evidence must state the soft-cap threshold so it reads as a heuristic');
|
|
} finally {
|
|
await rm(home, { recursive: true, force: true });
|
|
}
|
|
});
|
|
|
|
it('recommends trimming the description toward its trigger phrases', async () => {
|
|
const home = await homeWithOneAgent(PER_AGENT_DESC_SOFT_CAP + 100);
|
|
try {
|
|
const result = await runScannerWithHome(home);
|
|
const advisory = findPerAgent(result.findings)[0];
|
|
assert.ok(advisory, 'expected a per-agent advisory to carry remediation');
|
|
assert.match(String(advisory.recommendation), /trim|trigger|shorten/i,
|
|
'recommendation should point at trimming the description');
|
|
} finally {
|
|
await rm(home, { recursive: true, force: true });
|
|
}
|
|
});
|
|
|
|
it('emits per-agent advisories BEFORE the aggregate roll-up', async () => {
|
|
// 17 large agents → both per-agent advisories AND the aggregate fire.
|
|
const home = await homeWithNUserAgents(17, 1000);
|
|
try {
|
|
const result = await runScannerWithHome(home);
|
|
const agg = findAggregate(result.findings);
|
|
assert.ok(agg, 'expected the aggregate to fire alongside per-agent advisories');
|
|
const aggIndex = result.findings.indexOf(agg);
|
|
assert.equal(aggIndex, result.findings.length - 1,
|
|
'the aggregate roll-up must be emitted last (per-agent advisories precede it, mirroring SKL 001→002)');
|
|
assert.equal(findPerAgent(result.findings).length, 17,
|
|
'each of the 17 over-cap agents should carry its own advisory');
|
|
} finally {
|
|
await rm(home, { recursive: true, force: true });
|
|
}
|
|
});
|
|
});
|