Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion plugins-claude/agentic-ide/.claude-plugin/plugin.json
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
{
"name": "agentic-ide",
"version": "1.0.3",
"version": "1.0.5",
"description": "IDE-grade code intelligence for agents — Serena (LSP symbol nav/refactor), ast-grep (AST search/rewrite), and Semgrep (security & dataflow) bundled with usage cheatsheets, setup helpers, and a context-isolated explorer agent",
"author": {
"name": "Logan Gagne"
Expand Down
10 changes: 2 additions & 8 deletions plugins-claude/agentic-ide/skills/setup/SKILL.md
Original file line number Diff line number Diff line change
Expand Up @@ -67,13 +67,7 @@ Register in `~/.claude.json` (user-level `mcpServers`, or per-project under `.pr
"serena": {
"type": "stdio",
"command": "serena",
"args": [
"start-mcp-server",
"--context", "claude-code",
"--project-from-cwd",
"--disable-memories",
"--disable-onboarding"
],
"args": ["start-mcp-server", "--context", "claude-code", "--project-from-cwd"],
"env": {}
}
}
Expand All @@ -82,7 +76,7 @@ Register in `~/.claude.json` (user-level `mcpServers`, or per-project under `.pr

- `--context claude-code` tunes prompts and tool descriptions.
- `--project-from-cwd` auto-detects the project — no per-project config needed.
- `--disable-memories --disable-onboarding` skips Serena's opaque sidecar notes; project context belongs in source-controlled files.
- Create `~/.serena/serena_config.yml` with `base_modes: [no-memories]` to disable memories and onboarding tools globally. If you need a one-off invocation instead, pass `--mode interactive --mode editing --mode no-memories` to `serena start-mcp-server` and re-list any other modes you want.

Reconnect Claude Code. `mcp__serena__*` tools should appear; `mcp__serena__get_symbols_overview` on a source file should return structured data. Logs at `~/.serena/logs/`.

Expand Down
2 changes: 1 addition & 1 deletion plugins-copilot/agentic-ide/.claude-plugin/plugin.json
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
{
"name": "agentic-ide",
"version": "1.0.3",
"version": "1.0.5",
"description": "IDE-grade code intelligence for agents — Serena (LSP symbol nav/refactor), ast-grep (AST search/rewrite), and Semgrep (security & dataflow) bundled with usage cheatsheets, setup helpers, and a context-isolated explorer agent",
"author": {
"name": "Logan Gagne"
Expand Down
49 changes: 9 additions & 40 deletions plugins-copilot/agentic-ide/commands/setup.md
Original file line number Diff line number Diff line change
Expand Up @@ -58,37 +58,22 @@ uv tool install --from git+https://github.com/oraios/serena serena

Verify: `which serena && serena --version`. Upgrade later with `uv tool upgrade serena`.

Register the server with Copilot CLI. Quickest path — the `copilot mcp add` subcommand:
Create the Serena mode config and register the server with Copilot CLI:

```bash
copilot mcp add serena -- serena start-mcp-server \
--context=copilot-cli --project-from-cwd --disable-memories --disable-onboarding
```
mkdir -p ~/.serena

cat > ~/.serena/serena_config.yml <<'YAML'
base_modes:
- no-memories
YAML
Comment on lines +66 to +69

Or add it to `~/.copilot/mcp-config.json` (user-level; project-level alternative: `.mcp.json` at the repo root):

```json
{
"mcpServers": {
"serena": {
"type": "stdio",
"command": "serena",
"args": [
"start-mcp-server",
"--context=copilot-cli",
"--project-from-cwd",
"--disable-memories",
"--disable-onboarding"
],
"tools": ["*"]
}
}
}
copilot mcp add serena -- serena start-mcp-server --context=copilot-cli --project-from-cwd
```

- `--context=copilot-cli` tunes prompts and tool descriptions for Copilot CLI.
- `--project-from-cwd` auto-detects the project — no per-project config needed.
- `--disable-memories --disable-onboarding` skips Serena's opaque sidecar notes; project context belongs in source-controlled files.
- `~/.serena/serena_config.yml` with `base_modes: [no-memories]` disables memories and onboarding tools globally. If you need a one-off invocation instead, pass `--mode interactive --mode editing --mode no-memories` to `serena start-mcp-server` and re-list any other modes you want.

Restart the Copilot CLI session. `serena-*` tools should appear; `serena-get_symbols_overview` on a source file should return structured data. Logs at `~/.serena/logs/`.

Expand Down Expand Up @@ -123,22 +108,6 @@ Register the server with Copilot CLI — via `copilot mcp add`:
copilot mcp add semgrep -- semgrep-mcp
```

Or add it to `~/.copilot/mcp-config.json`:

```json
{
"mcpServers": {
"semgrep": {
"type": "stdio",
"command": "semgrep-mcp",
"args": [],
"tools": ["*"],
"env": {}
}
}
}
```

Optional: add `SEMGREP_APP_TOKEN` to `env` to enable `semgrep_findings` (pulls from Semgrep AppSec Platform). Generate at <https://semgrep.dev/orgs/-/settings/tokens>. All local-scan tools work without it.

Comment on lines 109 to 112
Restart the Copilot CLI session. `semgrep-*` tools should appear; `semgrep-supported_languages` should return a language list.
Expand Down
Loading