Detect NestJS projects and emit `route` nodes (each linked by a `references`
edge to its handler method) across all four transport layers:
- HTTP controllers: @Controller prefix joined with
@Get/@Post/@Put/@Patch/@Delete/@Head/@Options/@All
- GraphQL resolvers: @Query/@Mutation/@Subscription
- Microservices: @MessagePattern/@EventPattern
- WebSocket gateways: @SubscribeMessage (prefixed with gateway namespace)
Detected from any @nestjs/* dependency in package.json (falls back to scanning
*.controller.ts/*.resolver.ts/*.gateway.ts). Handles class+method path joining
with empty @Controller()/@Get(), a string-aware balanced-paren arg reader so
GraphQL type thunks (@Query(() => [User])) aren't truncated, stacked decorators
(@UseGuards) when locating the handler, and disambiguates the @Query() GraphQL
method decorator from the REST @Query() param decorator (GraphQL only counts
inside @Resolver classes). Also resolves injected *Service/*Controller refs to
their classes by Nest file-naming convention.
Adds 18 framework tests; updates the README framework table and CHANGELOG.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* perf(mcp): steer agents to answer directly instead of delegating to subagents
CodeGraph beats native grep/read on cost only when the agent queries it
directly. When the agent delegates to file-reading sub-agents, those
sub-agents read files regardless of the index, so CodeGraph becomes net
overhead on top of the reads. The install templates even told agents to
"spawn a subagent for explore-class questions" — the expensive path.
Changes:
- server-instructions + both install templates: add an "Answer directly —
don't delegate exploration" directive; reposition codegraph_explore as the
efficient one-call multi-symbol tool (was: "spawn a subagent for it").
- codegraph_explore: hard-cap output to its adaptive budget (it overran,
~30k vs a 28k cap) and tighten the medium tier (28k->13k).
- codegraph_node: return a member outline for container kinds instead of the
full class body.
Rigorous N>=4-per-arm warm-block benchmark (median total_cost_usd):
excalidraw (~600 files): WITH $0.54 vs native $1.02 (-47%)
vscode (~10k files): WITH $0.41 vs native $0.72 (-42%)
ky (~25 files): WITH $0.46 vs native $0.44 (wash)
Answers were equal-or-better (correct, file:line-cited) with ~6x fewer tool
calls; the directive drove the direct path on 14/14 codegraph runs.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* docs(readme): rebuild benchmark with real-world repos + cost/token/time/tool savings
Replace the "Claude Code (Python+Rust/Java)" rows — which benchmarked the
Claude Code CLI repo, not real codebases in those languages — with real
open-source projects per language: Django (Python), Tokio (Rust), OkHttp
(Java), Gin (Go), plus Alamofire (Swift) and the existing TypeScript repos
(VS Code, Excalidraw).
The table now reports all four savings the change targets — cost, tokens,
time, tool calls — as the median of 4 runs per arm (Claude Opus 4.7,
headless claude -p, with vs empty MCP config). Averages across the 7 repos:
35% cheaper, 59% fewer tokens, 49% faster, 70% fewer tool calls. Adds a
methodology note and raw WITH->WITHOUT medians.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Both git fast-paths in ExtractionOrchestrator (sync and getChangedFiles)
classified every untracked (`??`) file as "added" without checking the
index. Indexing a file doesn't make git track it, so the file stayed `??`
and was re-reported as pending and re-indexed on every run: `codegraph
status` listed it under Pending Changes forever and each `sync` re-added
it, even though its symbols were already queryable.
Merge the modified + added handling into a single hash-compared loop so
untracked files get the same treatment as tracked ones: "added" only if
missing from the index, "modified" if contents changed, skipped otherwise.
The non-git fallback path already did this and is unchanged.
Closes#206. Reported by @15290391025.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
`codegraph init -i` from a git super-repo containing independent nested
git repositories (not submodules) reported "No files found to index":
git ls-files reports an embedded repo only as an opaque `subdir/` entry
and never lists its files. Detect embedded repos via that trailing-slash
signal and recurse `git ls-files` into each, indexing tracked + untracked
source and honoring each repo's own .gitignore.
Reported by @timxx.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
better-sqlite3 ^11.0.0 (latest 11.10.0) ships no prebuilt binary for
Node 24's ABI (node-v137) and predates Node 24, so every Node 24 install
silently fell back to the 5-10x-slower WASM backend. Bump to ^12.4.1 —
the first 12.x with the Node 24 prebuild — and raise the engines floor to
Node 20 (Node 18 is EOL and dropped from better-sqlite3 12.x prebuilds).
Verified on macOS Node 24.15.0 (ABI 137): prebuilt binary used with no
compiler (installs even with CC/CXX sabotaged), `codegraph init -i` shows
no WASM banner, and `codegraph status` reports Backend: native. 639/639
tests pass on Node 22.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
MCP tools failed with "CodeGraph not initialized" when a client launched
the server outside the project and sent no rootUri/workspaceFolders — the
server fell back to its own cwd, missed the project's .codegraph/, and
returned a misleading "run codegraph init" error on every call. The only
workaround was passing projectPath by hand to each tool.
When no explicit path is given, the server now asks the client for its
workspace root via the standard MCP roots/list request (gated on the
client advertising the roots capability) before falling back to cwd. This
required teaching the stdio transport to send server->client requests and
match their responses by id (previously responses were dropped as invalid).
When a project still can't be resolved, the error now names the directory
it searched and tells the user to pass projectPath or add --path to the
MCP config, instead of pointing at a re-init they don't need.
Reported-by: @zhangyu1197
Closes#196
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Recursive fs.watch on a WSL2 /mnt NTFS/9p mount walks the directory tree
with every readdir/stat crossing the Windows boundary, stalling the event
loop long enough to blow past opencode's 30s MCP handshake timeout so the
tools never appear. This is the file-watcher half of the #172 fix, which
moved the DB/WASM open off the handshake but left the watcher on the
critical path.
- Add watchDisabledReason() policy: CODEGRAPH_NO_WATCH (off) >
CODEGRAPH_FORCE_WATCH (force on) > WSL2 + /mnt auto-detect (off).
FileWatcher.start() and the MCP server both honor it; the server now
logs why watching is off and how to refresh.
- Add `codegraph serve --mcp --no-watch`.
- When watching is off, init/install offer git sync hooks (post-commit,
post-merge, post-checkout) that run `codegraph sync` in the background,
or fall back to manual sync; either way the user is told the index
stays frozen until re-synced. uninit removes the hooks.
- Tests: watch-policy + git-hooks (idempotency, user-content preservation,
core.hooksPath).
Root-cause analysis and workaround by @mengfanbo123.
Closes#199
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Project-local installs wrote the MCP server to ./.claude.json, which Claude Code never reads — project-scoped servers must live in .mcp.json. The codegraph tools silently never loaded until users renamed the file by hand. Local installs now write ./.mcp.json and migrate any stale ./.claude.json entry on install and uninstall (siblings preserved). Global installs (~/.claude.json, user scope) were already correct.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* feat(mcp): line numbers in explore output + per-file cluster fixes
Follow-up to #185. Three changes to codegraph_explore:
1. Source sections now carry cat -n style line-number prefixes
(<num>\t<code>), so the agent can cite file:line straight from the
payload instead of re-Reading the file just to recover a line number.
Isolated A/B: the no-line-numbers arm spent 2 Reads + a grep to find a
line number the line-numbered arm cited with zero follow-up calls.
Payload cost ~3-5%. Toggle off with CODEGRAPH_EXPLORE_LINENUMS=0.
2. Per-file cluster selection now ranks clusters containing a query entry
point ahead of dense declaration blocks. Density-only ranking buried
the relevant methods (perform/didCreateURLRequest/task in Alamofire's
Session.swift) under the top-of-file class header + property list.
3. Whole-file "envelope" nodes (a class/struct/etc. spanning >50% of the
file) are excluded from clustering. The Session class spans ~1,400
lines; keeping it collapsed every method into one giant cluster that
tail-trimmed down to just the class header, hiding the methods.
Net vs the 0.7.10 baseline, line numbers on: Alamofire -60%, Excalidraw
-32%, VS Code -12% per explore call.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(mcp): language-neutral omission markers in explore output
The gap separator and the two tail-trim markers used C-style `//`
comments, which aren't comments in Python, Ruby, etc. Switch to plain
`... (gap) ...` / `... (trimmed) ...` so they read correctly inside any
language's fenced source block. With line numbers on, the line-number
jump already corroborates a gap.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(mcp): language-neutral truncation marker in codegraph_context
Sibling to the explore marker fix: codegraph_context's code-block
truncation used a C-style `// ... truncated ...`. Switch to
`... (truncated) ...` so it reads correctly in any language's fenced
source block.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* chore(release): bump version to 0.7.11
---------
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Output is now scaled to indexed file count. Small projects (<500 files)
cap at ~18KB and skip the "Additional relevant files" / completeness /
explore-budget reminders that earn their keep on larger codebases; medium
(<5,000) caps at ~28KB; large (<15,000) keeps the historical ~35KB; very
large goes up to ~38KB.
A per-file char cap also prevents a single file with many adjacent
symbols from collapsing into one whole-file dump (the pathological
Alamofire `Session.swift` case reported in #185), and a per-file symbol-
list cap stops the `#### path — sym(kind), ...` header from leaking
multi-KB lists when many adjacent symbols cluster together.
Measured against the README's benchmark repos: Alamofire (~100 files)
~62% smaller per call, Excalidraw (~600 files) ~35%, VS Code (~10k
files) ~14%. Agent-trust floor preserved — Relationships, scored cluster
selection, and structured-source output are all retained.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
GitHub renders release-note Markdown with GFM hard breaks, so every
`\n` becomes `<br>`. The CHANGELOG is hard-wrapped at ~75 chars for
readable diffs, which renders as awkward visible line breaks on the
release page (see https://github.com/colbymchenry/codegraph/releases/tag/v0.7.10).
Add `scripts/extract-release-notes.mjs` to extract a version block
and join indented continuation lines into a single line per bullet.
Nested list items, headings, and link references are preserved.
`scripts/release.sh` now uses this helper instead of the inline awk
extractor — repo-level CHANGELOG.md viewing is unaffected because
CommonMark there treats newlines as spaces.
Also fix the 0.7.10 entry: "Two underlying fixes" -> "Three", "Rust
file-/level" broken hyphen, and move the closes/credit line above
the nested list so it doesn't strand as a top-level paragraph.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
`codegraph_callees stage_apply::run` (and `_node`, `_impact`, ...)
returned "not found" against a repo with 7-9 sibling Rust modules,
each exporting `pub async fn run`. Two underlying issues:
1. The FTS5 query builder stripped `:` as a special char without
splitting on `::`, so `stage_apply::run` collapsed to the literal
`stage_applyrun` which matches nothing. Treat `::` as whitespace
before the strip step so both halves become FTS tokens.
2. `matchesSymbol` only understood `Parent.child` qualifiers and
relied on `qualifiedName` carrying the module path. Rust file-
level functions don't have their module name in `qualifiedName`
(it's encoded in the file path instead), so even dot-style
lookups failed. Accept `::`, `.`, `/` as separators; multi-level
forms compose; Rust `crate::`/`super::`/`self::` prefixes get
stripped before path matching. Fall back to file-path containment
when the qualified-name suffix doesn't match — `stage_apply::run`
matches a `run` in any file whose path has a `stage_apply` segment.
Also tightens the no-match branch: qualified lookups no longer fall
through to a fuzzy text match. `stage_apply::nonexistent_fn` returns
`null` instead of silently resolving to an unrelated `rollback` in
the same file.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The shimmer progress renderer writes from a worker thread via
`fs.writeSync(1, ...)` to keep the animation smooth while the main
thread is busy in SQLite. That path bypasses Node's TTY-aware
UTF-8->codepage conversion on Windows, so glyphs like `|`/`<>`/`-`
were emitted as raw UTF-8 bytes and reinterpreted by the console's
OEM codepage (CP437, CP936, ...), producing strings like
`鋍?[0m 鉒?[0m Scanning files 鈥?N found`.
Add `src/ui/glyphs.ts` with `supportsUnicode()` detection plus
matched Unicode + ASCII glyph sets, and route all CLI/shimmer
output through `getGlyphs()`. Defaults: ASCII on Windows and on
Linux kernel consoles (`TERM=linux`), Unicode everywhere else.
`CODEGRAPH_UNICODE=1` and `CODEGRAPH_ASCII=1` are escape hatches.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The MCP `initialize` handler was awaiting `tryInitializeDefault` —
which opens the SQLite DB and runs `await initGrammars()` (tree-sitter
WASM bootstrap) — before sending the JSON-RPC response. On slow
filesystems (Docker Desktop VirtioFS on macOS, WSL2) this could exceed
Claude Code's ~30s handshake timeout, leaving the codegraph child
process alive and unresponsive with no tools visible in the client.
Send the response first; defer the open to a tracked background
promise. The lazy retry path used by `tools/list` and `tools/call`
now awaits that promise instead of racing it with `openSync`, so we
never double-open the SQLite file.
Adds a subprocess-based regression test that asserts the JSON-RPC
response arrives on stdout before `startWatching()` logs to stderr.
This ordering check catches the regression on any filesystem, not
just slow ones where the timing matters in practice.
Reported by @sashanclrp; isolated by @sgrimm's wire capture.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* release: 0.7.7 (multi-agent installer — Cursor, Codex, opencode)
* fix(installer): opencode .jsonc + AGENTS.md (0.7.8)
v0.7.7 wrote ~/.config/opencode/opencode.json, but opencode reads
opencode.jsonc by default — so the codegraph MCP entry never appeared
in any opencode session. Also installs AGENTS.md so opencode's model
reaches for codegraph_* tools instead of native Grep.
- Prefer existing .jsonc, fall back to .json, default new installs
to .jsonc.
- Surgical edits via jsonc-parser preserve user comments and
formatting across install / re-install / uninstall round-trips.
- Install AGENTS.md (global ~/.config/opencode/AGENTS.md, local
./AGENTS.md) with the shared INSTRUCTIONS_TEMPLATE — same
marker-delimited approach Codex uses.
- +9 opencode-specific tests covering filename precedence, comment
preservation, AGENTS.md install + sibling-content preservation,
uninstall reverses both files.
575/575 tests pass. Hand-verified end-to-end: opencode session calls
codegraph_node + codegraph_callers for a structural query, zero Grep
calls.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* docs: overhaul CLAUDE.md and add scripts/release.sh + Cursor rules file
Replaces the old Claude-only CLAUDE.md with a comprehensive guide covering
the full project architecture, multi-agent installer, test conventions,
NodeKind/EdgeKind reference, and release workflow. Key additions:
- Documents the layered pipeline, all module paths, and the multi-target
installer (targets/, registry.ts, AgentTarget interface).
- Adds the Cursor `--path` quirk and the "update all three surfaces" rule
when changing MCP tool guidance.
- Documents `npm run eval`, `test:eval`, and the full set of build/test
commands including single-file patterns.
- `scripts/release.sh` — idempotent bash script that tags the current
commit, pushes the tag, and creates a GitHub Release whose notes are
extracted from the matching `## [X.Y.Z]` block in CHANGELOG.md. Safe
to re-run after partial failure.
- `.cursor/rules/codegraph.mdc` — Cursor-specific agent instructions
(tool decision table, rules of thumb, index-lag warning) written by
the installer and kept in sync with server-instructions.ts and
instructions-template.ts.
---------
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The 0.7.5 tarball shipped `dist/bin/codegraph.js` without the executable
bit set, causing `zsh: permission denied: codegraph` after a fresh global
install. The build script now `chmod +x`'s the binary before packing.
Also adds CHANGELOG.md and documents the release workflow in CLAUDE.md.
