Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Goland Doctor

ASecurity

One-command health check for the GoLand MCP oracle: is it connected, is the right project bound, and is the project model loaded correctly (Go module synced)? Load when the user asks to check / diagnose / verify the GoLand (IDE / MCP / oracle) connection or status, or when an agent is about to rely on the IDE oracle's semantic results and needs to confirm they are trustworthy.

16 stars
0 votes
0 copies
0 views
Added 9/20/2026
ai-agentsrustgogitsecurity

Works with

claude codemcp

Security Analysis

A100/100

Scanned 9/20/2026

Install to Claude Code

$npx -y skills add woditschka/agentic-coding-reference --skill goland-doctor --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Goland Doctor?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Goland Doctor
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/woditschka-goland-doctor-086b9ee7/badge)](https://www.skillsdirectory.com/skills/woditschka-goland-doctor-086b9ee7)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: goland-doctor
description: >-
  One-command health check for the GoLand MCP oracle: is it connected, is the
  right project bound, and is the project model loaded correctly (Go module
  synced)? Load when the user asks to check / diagnose / verify the GoLand (IDE /
  MCP / oracle) connection or status, or when an agent is about to rely on the
  IDE oracle's semantic results and needs to confirm they are trustworthy.
compatibility:
  - claude-code
  - github-copilot
metadata:
  version: "1.1"
  author: team
---

# goland-doctor — diagnose the GoLand MCP oracle

A responding MCP server is **not** the same as a usable oracle. The tools can answer while bound to a project that never synced its Go module, or to a *different* project than the one you are working in — and then `search_symbol` returns nothing and `get_file_problems` emits false "Unresolved reference" errors on code that `go build ./...` compiles cleanly. This skill runs the minimal probe set that tells the three states apart and prints a plain verdict.

For the *why* behind each probe and the fallback rules, see the `goland` skill (§ Connection health check). This skill is the **runner**; that skill is the **reference** — keep interpretation logic in sync with it.

## Procedure

`PROJECT=<absolute path to this repository's root>` (pass as `projectPath` on every call). Tool names below are bare — call each as it appears in your own tool list.

Run the probes your agent actually has (see § Tool availability per agent). `search_symbol` is the **portable primary** — every IDE-enabled agent has it, and it alone distinguishes all the states. `get_project_dependencies` is the **un-fakeable corroboration**, available only to the orchestrator and system-design-expert.

1. **Pick a known repo symbol** so the "right project" check is robust to renames. Glob `**/*.go` (skip `_test.go`), take any one file, and use an exported type or function name you can confirm exists. Note its real path — that is what a healthy oracle must echo back.

2. **Primary probe — symbol resolution (every IDE-enabled agent has this).** `search_symbol(q=<known symbol>, projectPath=PROJECT)`
   - Hit whose `filePath` is under this repo (matches step 1) → the **right project** is bound and its sources are indexed → healthy.
   - Hit whose `filePath` is *outside* this repo → **wrong project** bound.
   - `{"items":[]}` → sources not in the model: model not loaded (Go module not synced) or wrong project — disambiguate with step 3 if you can.
   - Call errors / times out → server **not connected**.

3. **Corroborating probe — dependencies (only if your agent declares it).** `get_project_dependencies(projectPath=PROJECT)`
   - Non-empty (the module's resolved `require` set) → Go module model synced; confirms a healthy `search_symbol`.
   - `{"dependencies":[]}` on a module that declares dependencies → **model not loaded** (no module graph). Use it to split a `[]` symbol result into *not-loaded* vs *wrong-project*. (A zero-dependency module is legitimately empty — corroborate with the symbol probe, not deps alone.)
   - If your agent cannot call this tool, say so and rely on `search_symbol` alone — never report a dependency count you did not observe.

4. **Do NOT use `get_project_modules` as a health signal.** A bare module lists fine on a broken model — a **false green**. Run it only for extra detail, never as the verdict.

## Tool availability per agent

The IDE tools are partitioned by role, so not every caller can run every probe. `search_symbol` (with `get_symbol_info`) is the common denominator — this skill is built around it for that reason.

| Caller | `search_symbol` (primary) | `get_project_dependencies` (corroboration) |
|---|:--:|:--:|
| Orchestrator (main loop) | ✅ | ✅ |
| system-design-expert | ✅ | ✅ |
| feature-implementer | ✅ | ❌ |
| code / test / security reviewers | ✅ | ❌ |
| doc-reviewer / coordinator / PRD | ❌ (no IDE tools) | ❌ |

A caller with no `search_symbol` at all cannot run this check — it reports "IDE not consulted (no oracle tools in this agent)" and proceeds on the native + `go` baseline. A caller with `search_symbol` but not `get_project_dependencies` runs the primary probe and marks the corroboration `n/a`.

## Verdict mapping (primary axis = `search_symbol`)

| `search_symbol(known symbol)` | `get_project_dependencies` (if available) | Verdict |
|---|---|---|
| call errors / times out | — | **NOT CONNECTED** |
| hit, `filePath` under this repo | populated *or* n/a | **HEALTHY** |
| hit, `filePath` outside this repo | — | **WRONG PROJECT BOUND** |
| `[]` | `[]` (module declares deps) | **CONNECTED — MODEL NOT LOADED** (Go module not synced) |
| `[]` | populated | **PARTIAL** (deps loaded but symbols unindexed — indexing in progress or stale; re-run shortly) |
| `[]` | n/a (agent lacks the tool) | **NOT USABLE** — symbols don't resolve and there's no deps probe to say why. Treat as unusable (fall back to grep + `go build`); have the orchestrator or system-design-expert re-run with `get_project_dependencies`, or check the IDE directly (§ Without Claude Code). |

A repo-local `search_symbol` hit is sufficient for **HEALTHY** on its own — the dependencies probe corroborates it but is not required. The deps probe earns its keep only on a `[]` symbol result, where it separates *not-loaded* from *unindexed*.

The `[]` / `[]` → **MODEL NOT LOADED** row holds only when the module declares dependencies. A legitimately dependency-free module (stdlib only) returns `[]` deps even when healthy, so on such a module ignore the deps probe and take the `search_symbol` verdict alone.

## Output format

Print this block verbatim, filled in:

```
GoLand MCP Oracle — Status
  Connection:    <connected | NOT CONNECTED>
  Project model: <loaded (module synced) | NOT LOADED | unknown>
  Right project: <yes — KnownSymbol resolves at <path> | WRONG project | cannot confirm>
  Verdict:       <HEALTHY | CONNECTED — MODEL NOT LOADED | WRONG PROJECT BOUND | PARTIAL | NOT USABLE | NOT CONNECTED>
  Probes:        search_symbol(<KnownSymbol>)=<M hits>, get_project_dependencies=<N libs | n/a for this agent>
  Trust oracle?  <YES — use at the chokepoints | NO — fall back to grep + go build>
  Remedy:        <none | the matching remedy below>
```

## Remedies

- **MODEL NOT LOADED** — in the IDE the MCP is attached to: *File → Reload Project* (or *File → Invalidate Caches / Restart*); confirm the **Go Modules** / **External Libraries** view populates and module mode is enabled. Then re-run `goland-doctor`.
- **WRONG PROJECT BOUND** — close the stray project/instance, or ensure only one IDE window is open for this repo; the MCP binds to the running IDE.
- **NOT CONNECTED** — confirm the IDE is running with the MCP server enabled (*Settings → Tools → MCP Server*, or the MCP plugin's status indicator). Do not block work on it — fall back to native + `go`.

## Without Claude Code (human-only check)

Anyone can sanity-check from the IDE directly, no agent needed:

1. **Connected?** GoLand *Settings → Tools → MCP Server* shows the MCP server running.
2. **Right project + loaded?** The Project view shows this repo, the **Go Modules** tool window lists this module, and **External Libraries** lists the resolved dependencies. If External Libraries is empty or the module is not listed, the project was opened as a plain folder — *Reload Project* (or reopen with Go modules enabled).

## Rules

- Read-only. This skill never edits files; it only probes and reports.
- Report only probes you actually ran (see `goland` § Report only checks you actually ran). Never narrate a green you did not observe.
- A `get_project_modules` success is **not** a passing verdict on its own — the verdict comes from symbol resolution (primary), corroborated by dependencies when the caller has that tool.

Attribution

woditschkawoditschka
View sourceMore from woditschka →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

Caveman

Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

1023331 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3331 votes

catchup

Recovers prior coding-agent session context by running `catchup <agent> --since-compact`, which extracts a clean summary of a previous Codex, Claude Code, Antigravity, OpenCode, or Pi Agent session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", or asks to recover/summarize a previous session before continuing. Do NOT use for the current conversation, git history, or any non-agent log.

611 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →