|
|
@@ -124,41 +124,52 @@ export interface ExploreOutputBudget {
|
|
|
includeCompletenessSignal: boolean;
|
|
|
/** Include the explore-budget reminder at the end. */
|
|
|
includeBudgetNote: boolean;
|
|
|
+ /**
|
|
|
+ * Hard-drop test/spec/icon/i18n files from the relevant-file set unless
|
|
|
+ * the query itself mentions tests. Today they're only deprioritized in
|
|
|
+ * the sort, which on tiny repos still lets one slip into the top N (e.g.
|
|
|
+ * cobra's `command_test.go` displaced `args.go` and contributed ~10KB of
|
|
|
+ * pure noise to "How does cobra parse commands?"). Off by default; on
|
|
|
+ * for the very-tiny tier where one slip dominates the budget.
|
|
|
+ */
|
|
|
+ excludeLowValueFiles: boolean;
|
|
|
}
|
|
|
|
|
|
export function getExploreOutputBudget(fileCount: number): ExploreOutputBudget {
|
|
|
if (fileCount < 150) {
|
|
|
return {
|
|
|
- // Very-tiny tier paired with the tool gating in ToolHandler.getTools
|
|
|
- // (<150 files exposes only 5 core tools). Together: ~50% prompt
|
|
|
- // overhead reduction + tighter explore output. Per-file kept at
|
|
|
- // 3800 (the next tier's value) to satisfy the monotonic invariant.
|
|
|
- // Relationships kept ON — cheap structural signal that survives
|
|
|
- // even after the budget cut.
|
|
|
+ // ITER3: revert iter2's aggressive body shrink (forced Read fallback —
|
|
|
+ // the per-file 2.5K cap pushed the agent to Read instead of node).
|
|
|
+ // Back to the iter1 shape (13K/4/3.8K) but keep the test-file
|
|
|
+ // hard-exclude. The cost lever for this tier lives in handleContext
|
|
|
+ // (steering the agent to stop after 1-2 calls), not in this budget.
|
|
|
maxOutputChars: 13000,
|
|
|
defaultMaxFiles: 4,
|
|
|
maxCharsPerFile: 3800,
|
|
|
gapThreshold: 7,
|
|
|
maxSymbolsInFileHeader: 5,
|
|
|
maxEdgesPerRelationshipKind: 4,
|
|
|
- includeRelationships: true,
|
|
|
+ includeRelationships: false,
|
|
|
includeAdditionalFiles: false,
|
|
|
includeCompletenessSignal: false,
|
|
|
includeBudgetNote: false,
|
|
|
+ excludeLowValueFiles: true,
|
|
|
};
|
|
|
}
|
|
|
if (fileCount < 500) {
|
|
|
return {
|
|
|
+ // ITER3: same revert/keep-filter pattern as <150.
|
|
|
maxOutputChars: 18000,
|
|
|
defaultMaxFiles: 5,
|
|
|
maxCharsPerFile: 3800,
|
|
|
gapThreshold: 8,
|
|
|
maxSymbolsInFileHeader: 6,
|
|
|
maxEdgesPerRelationshipKind: 6,
|
|
|
- includeRelationships: true,
|
|
|
+ includeRelationships: false,
|
|
|
includeAdditionalFiles: false,
|
|
|
includeCompletenessSignal: false,
|
|
|
includeBudgetNote: false,
|
|
|
+ excludeLowValueFiles: true,
|
|
|
};
|
|
|
}
|
|
|
if (fileCount < 5000) {
|
|
|
@@ -178,6 +189,7 @@ export function getExploreOutputBudget(fileCount: number): ExploreOutputBudget {
|
|
|
includeAdditionalFiles: true,
|
|
|
includeCompletenessSignal: true,
|
|
|
includeBudgetNote: true,
|
|
|
+ excludeLowValueFiles: false,
|
|
|
};
|
|
|
}
|
|
|
if (fileCount < 15000) {
|
|
|
@@ -192,6 +204,7 @@ export function getExploreOutputBudget(fileCount: number): ExploreOutputBudget {
|
|
|
includeAdditionalFiles: true,
|
|
|
includeCompletenessSignal: true,
|
|
|
includeBudgetNote: true,
|
|
|
+ excludeLowValueFiles: false,
|
|
|
};
|
|
|
}
|
|
|
return {
|
|
|
@@ -205,6 +218,7 @@ export function getExploreOutputBudget(fileCount: number): ExploreOutputBudget {
|
|
|
includeAdditionalFiles: true,
|
|
|
includeCompletenessSignal: true,
|
|
|
includeBudgetNote: true,
|
|
|
+ excludeLowValueFiles: false,
|
|
|
};
|
|
|
}
|
|
|
|
|
|
@@ -688,7 +702,13 @@ export class ToolHandler {
|
|
|
// 5 is the empirical lower bound. Tools beyond search/context/
|
|
|
// node/explore/trace pay overhead that the agent doesn't recoup
|
|
|
// on tiny-repo flow questions.
|
|
|
- const TINY_REPO_FILE_THRESHOLD = 150;
|
|
|
+ // ITER4: raise threshold 150 → 500 so single-file frameworks
|
|
|
+ // (sinatra at 159, slim_framework around 200) also get the
|
|
|
+ // 5-tool surface. The empirical 5-tool floor was set on <150
|
|
|
+ // probes; iter3 measurement showed sinatra is structurally the
|
|
|
+ // SAME problem as cobra (single-file WITHOUT-arm Read wins),
|
|
|
+ // so it deserves the same gating.
|
|
|
+ const TINY_REPO_FILE_THRESHOLD = 500;
|
|
|
const TINY_REPO_CORE_TOOLS = new Set([
|
|
|
'codegraph_search',
|
|
|
'codegraph_context',
|
|
|
@@ -1095,9 +1115,12 @@ export class ToolHandler {
|
|
|
// 8 covers the typical 1-3 entry-point + their immediate neighbors
|
|
|
// without dragging in the rest of the small codebase.
|
|
|
let defaultMaxNodes = 20;
|
|
|
+ let isTinyRepo = false;
|
|
|
+ let isSmallRepo = false;
|
|
|
try {
|
|
|
const stats = cg.getStats();
|
|
|
- if (stats.fileCount < 150) defaultMaxNodes = 8;
|
|
|
+ if (stats.fileCount < 150) { defaultMaxNodes = 8; isTinyRepo = true; }
|
|
|
+ else if (stats.fileCount < 500) { isSmallRepo = true; }
|
|
|
} catch {
|
|
|
// stats failure — fall back to the standard default
|
|
|
}
|
|
|
@@ -1123,13 +1146,39 @@ export class ToolHandler {
|
|
|
// multi-module flow questions (Q3 / etcd Q2 in the audit).
|
|
|
const flowTrace = await this.maybeInlineFlowTrace(task, cg);
|
|
|
|
|
|
+ // Iter3 — sufficiency steering on small repos.
|
|
|
+ //
|
|
|
+ // Measured economics on tiny (<150) and small (<500) projects: every
|
|
|
+ // additional MCP tool call costs ~$0.02-0.05 in cache-write tokens
|
|
|
+ // (5K-15K per response at $3.75/1M). The agent reflexively follows
|
|
|
+ // codegraph_context with explore/node even when the context response
|
|
|
+ // is already sufficient — that pattern drove the cost gap that
|
|
|
+ // smaller bodies (iter2) failed to close (smaller bodies just shifted
|
|
|
+ // the agent to Read instead). Direct directive on small-repo
|
|
|
+ // responses: tell the agent the context call IS the comprehensive
|
|
|
+ // pass for a project of this size and that follow-ups should be
|
|
|
+ // narrow (trace from→to, node single-symbol) — not another broad
|
|
|
+ // explore that re-bundles the same content.
|
|
|
+ // ITER4: unified strong directive for both tiny (<150) and small
|
|
|
+ // (<500) tiers — measured iter3 result was that the soft <500
|
|
|
+ // wording was IGNORED on sinatra (5 tool calls, +92% loss) while
|
|
|
+ // the strong <150 wording was followed on cobra/slim (3 calls,
|
|
|
+ // -21%/-22% wins). The single-file-framework problem (sinatra)
|
|
|
+ // is structurally the same as cobra's; both deserve the same
|
|
|
+ // sufficiency steering.
|
|
|
+ let smallRepoTail = '';
|
|
|
+ if (isTinyRepo || isSmallRepo) {
|
|
|
+ const sizeQualifier = isTinyRepo ? 'under 150' : 'under 500';
|
|
|
+ smallRepoTail = `\n\n---\n> **This project is small** (${sizeQualifier} indexed files). The entry points and code above cover the relevant surface — **do NOT call codegraph_explore as a follow-up; its content will largely duplicate this response**. If you need a specific flow, call \`codegraph_trace from→to\`. If you need one specific symbol's body, call \`codegraph_node <name>\`. Otherwise, answer from what is above.`;
|
|
|
+ }
|
|
|
+
|
|
|
// buildContext returns string when format is 'markdown'
|
|
|
if (typeof context === 'string') {
|
|
|
- return this.textResult(this.truncateOutput(context + flowTrace + reminder));
|
|
|
+ return this.textResult(this.truncateOutput(context + flowTrace + reminder + smallRepoTail));
|
|
|
}
|
|
|
|
|
|
// If it returns TaskContext, format it
|
|
|
- return this.textResult(this.truncateOutput(this.formatTaskContext(context) + flowTrace + reminder));
|
|
|
+ return this.textResult(this.truncateOutput(this.formatTaskContext(context) + flowTrace + reminder + smallRepoTail));
|
|
|
}
|
|
|
|
|
|
/**
|
|
|
@@ -1176,6 +1225,7 @@ export class ToolHandler {
|
|
|
seen.add(key);
|
|
|
ids.push(sym);
|
|
|
}
|
|
|
+
|
|
|
if (ids.length < 2) return '';
|
|
|
|
|
|
// The first two distinct symbols, in order of appearance, are the most
|
|
|
@@ -1950,11 +2000,52 @@ export class ToolHandler {
|
|
|
}
|
|
|
|
|
|
// Only include files that have entry points or nodes directly connected to entry points
|
|
|
- const relevantFiles = [...fileGroups.entries()].filter(([, group]) => group.score >= 3);
|
|
|
+ let relevantFiles = [...fileGroups.entries()].filter(([, group]) => group.score >= 3);
|
|
|
|
|
|
// Extract query terms for relevance checking
|
|
|
const queryTerms = query.toLowerCase().split(/\s+/).filter(t => t.length >= 3);
|
|
|
|
|
|
+ // Test/spec/icon/i18n file detector — used both for the pre-sort hard
|
|
|
+ // filter (tiny tier) and the comparator deprioritization (all tiers).
|
|
|
+ const isLowValue = (p: string) => {
|
|
|
+ const lp = p.toLowerCase();
|
|
|
+ return (
|
|
|
+ /\/(tests?|__tests?__|spec)\//.test(lp) ||
|
|
|
+ /_test\.go$/.test(lp) ||
|
|
|
+ /(?:^|\/)test_[^/]+\.py$/.test(lp) ||
|
|
|
+ /_test\.py$/.test(lp) ||
|
|
|
+ /_spec\.rb$/.test(lp) ||
|
|
|
+ /_test\.rb$/.test(lp) ||
|
|
|
+ /\.(test|spec)\.[jt]sx?$/.test(lp) ||
|
|
|
+ /(test|spec|tests)\.(java|kt|scala)$/.test(lp) ||
|
|
|
+ /(tests?|spec)\.cs$/.test(lp) ||
|
|
|
+ /tests?\.swift$/.test(lp) ||
|
|
|
+ /_test\.dart$/.test(lp) ||
|
|
|
+ /\bicons?\b/.test(lp) ||
|
|
|
+ /\bi18n\b/.test(lp)
|
|
|
+ );
|
|
|
+ };
|
|
|
+
|
|
|
+ // Tiny-tier hard-exclude: on small projects (`excludeLowValueFiles`
|
|
|
+ // budget flag), one slipped test/spec file dominates the per-file budget
|
|
|
+ // (cobra's `command_test.go` displaced `args.go` and contributed ~10KB of
|
|
|
+ // pure noise to "How does cobra parse commands?"). The sort-step
|
|
|
+ // deprioritization isn't enough at small N. Skip the hard-exclude when
|
|
|
+ // the query itself is about tests — that's the legitimate "explore the
|
|
|
+ // tests" case where the agent does want them.
|
|
|
+ if (budget.excludeLowValueFiles) {
|
|
|
+ const queryMentionsTests = /\b(test|tests|testing|spec|verify|verifies)\b/i.test(query);
|
|
|
+ if (!queryMentionsTests) {
|
|
|
+ const nonLow = relevantFiles.filter(([p]) => !isLowValue(p));
|
|
|
+ // Only apply the hard-filter if we still have at least 2 non-test
|
|
|
+ // candidates after the cut — otherwise the agent is asking about an
|
|
|
+ // area where tests are the only signal, and we should not strip them.
|
|
|
+ if (nonLow.length >= 2) {
|
|
|
+ relevantFiles = nonLow;
|
|
|
+ }
|
|
|
+ }
|
|
|
+ }
|
|
|
+
|
|
|
// Sort files: highest relevance first, deprioritize low-value files
|
|
|
const sortedFiles = relevantFiles.sort((a, b) => {
|
|
|
const aPath = a[0].toLowerCase();
|
|
|
@@ -1971,36 +2062,6 @@ export class ToolHandler {
|
|
|
const bRelevant = hasQueryRelevance(bPath, b[1].nodes);
|
|
|
if (aRelevant !== bRelevant) return aRelevant ? -1 : 1;
|
|
|
|
|
|
- // Deprioritize test files, icon files, and i18n files. Covers both
|
|
|
- // directory-style (`/tests/`, `/spec/`) AND suffix-style conventions
|
|
|
- // across every language we support — without the suffix check, etcd's
|
|
|
- // `watchable_store_test.go` displaced 5K chars of real-flow source in
|
|
|
- // codegraph_explore for Q2.
|
|
|
- const isLowValue = (p: string) =>
|
|
|
- /\/(tests?|__tests?__|spec)\//i.test(p) ||
|
|
|
- // Go: `*_test.go`
|
|
|
- /_test\.go$/i.test(p) ||
|
|
|
- // Python: `test_*.py` (pytest discovery) and `*_test.py`
|
|
|
- /(?:^|\/)test_[^/]+\.py$/i.test(p) ||
|
|
|
- /_test\.py$/i.test(p) ||
|
|
|
- // Ruby: `*_spec.rb` (rspec) and `*_test.rb` (minitest)
|
|
|
- /_spec\.rb$/i.test(p) ||
|
|
|
- /_test\.rb$/i.test(p) ||
|
|
|
- // JS / TS: `*.test.ts`, `*.spec.tsx`, etc.
|
|
|
- /\.(test|spec)\.[jt]sx?$/i.test(p) ||
|
|
|
- // JVM: `*Test.java`, `*Tests.java`, `*Spec.kt`, `*Spec.scala`
|
|
|
- /(Test|Spec|Tests)\.(java|kt|scala)$/.test(p) ||
|
|
|
- // C#: `*Tests.cs`, `*Test.cs`, `*Spec.cs`
|
|
|
- /(Tests?|Spec)\.cs$/.test(p) ||
|
|
|
- // Swift: `*Tests.swift` (XCTest convention)
|
|
|
- /Tests?\.swift$/.test(p) ||
|
|
|
- // Dart: `*_test.dart`
|
|
|
- /_test\.dart$/i.test(p) ||
|
|
|
- // Rust: `tests/*.rs` already caught by `/tests/` above; `_test.rs`
|
|
|
- // and `_tests.rs` aren't Rust conventions (Rust uses `#[cfg(test)]`
|
|
|
- // inside source files), so nothing extra needed.
|
|
|
- /\bicons?\b/i.test(p) ||
|
|
|
- /\bi18n\b/i.test(p);
|
|
|
const aLow = isLowValue(aPath);
|
|
|
const bLow = isLowValue(bPath);
|
|
|
if (aLow !== bLow) return aLow ? 1 : -1;
|
Colby McHenry