Task-tool subagents never see the MCP initialize instructions and hold
the MCP tools only as deferred names they rarely think to load — so
delegated work bypassed codegraph almost entirely (measured ~1 of 9
forced-delegation runs touched it; the rest did 30-50 grep/read calls).
Two additions close the gap:
- CLI: `codegraph explore` and `codegraph node` call the same ToolHandler
as the MCP tools and print identical output — the graph for any agent
with a shell (subagents, Gemini CLI, raw Codex, humans).
- Installer: each agent target (claude/codex/gemini/opencode) writes a
short marker-fenced CodeGraph section into its instructions file —
the one channel subagents DO receive — naming both surfaces. Upsert
self-heals the stale pre-#529 long block; uninstall strips it; re-runs
are byte-equal unchanged. (#529's duplication argument bounded the
size: four lines, commands only.)
A/B (excalidraw, sonnet/high, forced Explore-agent delegation): without
the block, subagent codegraph usage ~1/9 runs; with it, 4/4 — subagents
ToolSearch-load the MCP tools and run explore 5-7x, best runs with ZERO
Read/grep (80-95s vs 150-197s baseline). The block's mechanism: the
parent relays the note into the task prompt, making the deferred tool
names salient.
Contract tests updated to the new expectations (write + self-heal
replace the #529 strip-only behavior); README install/guidance sections
refreshed (they also still described the pre-#817/#818 tool surface).
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
The installer wrote a `## CodeGraph` usage block into each agent's
instructions file (CLAUDE.md / AGENTS.md / GEMINI.md / .cursor/rules /
Kiro steering) that duplicated, almost verbatim, the guidance the MCP
server already emits in its `initialize` response — so agents that
surface MCP instructions (Claude Code) read the same playbook twice
every turn.
All 6 instruction-writing targets (claude, cursor, codex, opencode,
gemini, kiro) now stop writing the block. install self-heals by
stripping a block a previous version wrote (uninstall already did), so
the next `codegraph install`/`uninstall` cleans up existing installs;
upgrading the package alone does not (the leftover block is harmless).
server-instructions.ts is now the single source of truth — the two
steers unique to the old template ("trust codegraph, don't re-verify
with grep" and the not-initialized -> `init -i` hint) are ported there.
Removes the now-dead INSTRUCTIONS_TEMPLATE / CLAUDE_MD_TEMPLATE,
claude-md-template.ts, writeClaudeMd / hasClaudeMdSection, and the
Cursor-only wireProjectSurfaces bootstrap. The install log learned a
"Removed" verb. Tests rewritten to the new contract + self-heal
coverage (140/140 installer tests pass).
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
`codegraph install` now detects and configures two more agents:
- Gemini CLI / Antigravity CLI — `~/.gemini/settings.json` (or
`./.gemini/settings.json`) + `~/.gemini/GEMINI.md` (or project-root
`./GEMINI.md`). Preserves pre-existing top-level settings like
`security.auth` and sibling MCP servers.
- Antigravity IDE — writes to Antigravity's unified MCP config at
`~/.gemini/config/mcp_config.json` (post-migration, detected via
the `.migrated` marker Antigravity drops). Falls back to the
legacy `~/.gemini/antigravity/mcp_config.json` on pre-migration
builds; install migrates a stale legacy entry, uninstall sweeps
both. Antigravity-managed sibling fields (e.g. the `disabled` flag
added when users disable a server through the UI) survive re-install.
Two Antigravity-specific quirks the target handles:
1. Entries with `type: "stdio"` are silently rejected by
Antigravity's MCP scanner; we omit the field for this target.
2. macOS GUI apps launched from Dock/Finder get a stripped PATH
that excludes nvm — a bare `codegraph` command name fails to
spawn even when `which codegraph` works in the user's shell.
The target resolves `codegraph` to its absolute path at install
time on macOS. Linux + Windows are unaffected.
End-to-end validated:
- macOS: real Gemini CLI v0.43 via tmux — `/mcp` shows codegraph with
all 10 tools, `codegraph_status` executes and returns real index
state. Real Antigravity IDE shows codegraph under Customizations
after restart.
- Linux (Docker node:22-bookworm) + Windows (Parallels Win11): 116
installer tests pass; CLI install + uninstall round-trip verified.
Test coverage: the new targets inherit the existing parameterized
contract (idempotent install, sibling preservation, install/uninstall
round-trip). Plus 14 target-specific tests covering migration-marker
detection, legacy→unified entry migration, `disabled` flag
preservation, the `type` field omission, gemini+antigravity
coexistence in the same `~/.gemini/`, and macOS-only path resolution.
Full suite: 972 passing.
Closes#399.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
