Agent skill · Audited from source

Workspace

skill · alinaqi/claude-bootstrapMIT licenceAudited 14 Oct 2026 · source ea59a0c53d79

★ 707pushed today
✕BLOCKGrade D · 69/100 · 1 finding · ea59a0c
707repo stars · whole alinaqi/claude-bootstrap

Is it safe? No — BLOCK, grade D (69/100). We read this skill's source at commit ea59a0c53d79 and found no filesystem, network or shell access and no hard-coded secrets. The audit raised 1 finding, 1 critical or high. What it does: Dynamic multi-repo and monorepo awareness for Claude Code. Analyze workspace topology, track API contracts, and maintain cross-repo context.

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “Workspace (alinaqi/claude-bootstrap)”. Retrieved 14 Oct 2026. https://modelsatlas.com/skills/alinaqi/workspace

Blocked. The page stays public so you can check this skill if someone recommended it to you. The audit found 1 critical or high finding, starting with “Prompt injection” in SKILL.md:510. Do not install it without reading them.

Trust audit

Audit 0.4.1 · source read 14 Oct 2026

Trust score69/100✕BLOCK D
1 critical0 high0 medium0 low0 info

BLOCK means the audit found something that could harm you or your machine. Grade D is shared by 642 of 14,816 audited skills.

What the source does

  • FilesystemNone observedNo reads or writes in the source
  • NetworkNone observedNo outbound calls in the source
  • ShellNone observedNo commands run in the source
  • DependenciesPinnedNothing floats
  • SecretsNone foundNo keys in source

Five audit layers

LayerWhat we checkResultFindings
L0 ProvenanceRepo owner, licence, archive status, activityPASS0
L1 Static sourceSecrets, path traversal, supply chain, hidden codeNA0
L2 InstructionsPrompt injection, authority framing, hidden directivesFAIL1
L3 InventoryThe files and declarations match the sourcePASS0
L4 SandboxRun in isolation and watch behaviourSKIPPED why0

Gates applied: critical finding, no behavioural pass, undeclared transfer.

Findings (1)

CRITICALPrompt injectionprompt.transfer_instruction · SKILL.md:510
F-01 Prompt injectionCWE-94 · CWE-1427Layer L2
| Refresh token | POST /api/auth/refresh | apps/api | endpoint |
Why
an instruction to move sensitive data to an outside destination
Fix
remove; a skill never needs the user's secrets off the machine

Audit 0.4.1 · source sha ea59a0c53d79 · 14 Oct 2026Report an issue or request a re-scan

Provenance

Who published it, and what the verdict is pinned to

  • Publisheralinaqi GitHub userOwner of the repository this skill is read from
  • Source folderalinaqi/claude-bootstrap/skills/workspaceThe exact folder we audited
  • LicenceMITAs the repository states it
  • Commit pinned for this verdictea59a0c53d79A new audit runs when the source changes
  • Repository statusNot archivedPushed today
  • Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

What it tells the agent

Full source on GitHub ↗

Its description: Dynamic multi-repo and monorepo awareness for Claude Code. Analyze workspace topology, track API contracts, and maintain cross-repo context.

---
name: workspace
description: Dynamic multi-repo and monorepo awareness for Claude Code. Analyze workspace topology, track API contracts, and maintain cross-repo context.
when-to-use: When working across multiple repos or in a monorepo with shared dependencies
user-invocable: true
effort: high
---

# Workspace Analysis Skill

> Dynamic multi-repo and monorepo awareness for Claude Code. Analyze workspace topology, track API contracts, and maintain cross-repo context.

## The Problem

When you have separate frontend/backend repos (or monorepo with multiple apps), Claude Code operates in isolation. It doesn't know:

- API contracts between modules/repos
- Shared types and interfaces
- Full system architecture
- Cross-repo dependencies
- What changed in other parts of the system

This leads to:
- Duplicate type definitions
- API contract mismatches
- Breaking changes not caught until runtime
- Claude reimplementing things that exist elsewhere

---

## Solution: Dynamic Workspace Analysis

Instead of static manifests that get stale, Claude dynamically analyzes the workspace and generates context artifacts that stay fresh through hooks.

```
┌─────────────────────────────────────────────────────────────────┐
│  WORKSPACE ANALYSIS SYSTEM                                       │
├─────────────────────────────────────────────────────────────────┤
│                                                                  │
│  /analyze-workspace (Full Analysis - ~2 min)                     │
│  ├── Topology discovery (monorepo vs multi-repo)                │
│  ├── Dependency graph (who calls whom)                          │
│  ├── Contract extraction (OpenAPI, GraphQL, types)              │
│  └── Key file identification (what to load when)                │
│                                                                  │
│  /sync-contracts (Incremental - ~15 sec)                         │
│  ├── Check contract source files for changes                    │
│  ├── Update CONTRACTS.md with diffs                             │
│  └── Validate consistency                                       │
│                                                                  │
│  Hooks (Automatic)                                               │
│  ├── Session start: Staleness advisory (~5 sec)                 │
│  ├── Post-commit: Auto-sync if contracts changed (~15 sec)      │
│  └── Pre-push: Validation gate (~10 sec)                        │
│                                                                  │
└─────────────────────────────────────────────────────────────────┘
```

---

## Workspace Classification

### Detection Patterns

| Type | Indicators | File Access |
|------|------------|-------------|
| **Monorepo** | pnpm-workspace.yaml, nx.json, turbo.json, lerna.json | Direct (same tree) |
| **Multi-repo** | Sibling directories with separate .git | Via symlinks or paths |
| **Hybrid** | Monorepo + external repo dependencies | Mixed |
| **Single** | One app, no workspace config | N/A (use existing-repo) |

### Monorepo Detection

```bash
# Check for monorepo indicators
ls package.json pnpm-workspace.yaml lerna.json nx.json turbo.json 2>/dev/null
ls apps/ packages/ services/ libs/ modules/ 2>/dev/null
```

### Multi-Repo Detection

```bash
# Check sibling directories for related repos
ls -la ../*.git 2>/dev/null
cat ../*/.git/config 2>/dev/null | grep "url"

# Look for naming patterns
ls .. | grep -E "(frontend|backend|api|web|shared|common)"
```

### Polyglot Detection

```bash
# Find all package manifests
find . -maxdepth 4 -name "package.json" -o -name "pyproject.toml" \
  -o -name "go.mod" -o -name "Cargo.toml" -o -name "pom.xml" \
  -o -name "build.gradle" -o -name "Gemfile"
```

---

## Analysis Protocol

### Phase 1: Topology Discovery (~30 seconds)

Determine workspace structure:

```markdown
## Discovery Checklist

1. [ ] Identify workspace root
2. [ ] Classify workspace type (monorepo/multi-repo/hybrid/single)
3. [ ] List all modules/apps/packages
4. [ ] Detect tech stack per module
5. [ ] Identify entry points per module
```

**Module Detection Pattern:**

```
workspace-root/
├── apps/           → Application modules
│   ├── web/        → Frontend app
│   └── api/        → Backend app
├── packages/       → Shared packages
│   ├── ui/         → Component library
│   ├── types/      → Shared types
│   └── db/         → Database layer
├── services/       → Microservices
└── libs/           → Internal libraries
```

### Phase 2: Dependency Graph (~60 seconds)

For each module, map:

**1. Internal Dependencies**
```bash
# TypeScript/JavaScript
grep -r "from ['\"]@" --include="*.ts" --include="*.tsx" | head -50
grep -r "workspace:" package.json

# Python
grep -r "from \." --include="*.py" | head -50
```

**2. API Relationships**
```bash
# Find API calls
grep -rE "fetch|axios|httpx|requests\." --include="*.ts" --include="*.py" | \
  grep -E "/api|localhost|127\.0\.0\.1" | head -30
```

**3. Database Connections**
```bash
# Find DB access patterns
grep -rE "prisma|drizzle|sqlalchemy|sequelize|typeorm" --include="*.ts" --include="*.py"
```

### Phase 3: Contract Extraction (~45 seconds)

Identify and parse API contracts:

| Contract Type | Detection | Extraction |
|---------------|-----------|------------|
| **OpenAPI** | openapi.json, swagger.yaml, /docs endpoint | Parse paths, schemas |
| **GraphQL** | schema.graphql, *.gql, /graphql endpoint | Parse types, queries, mutations |
| **tRPC** | trpc router files, @trpc/* imports | Parse router definitions |
| **Protobuf** | *.proto files | Parse services, messages |
| **TypeScript** | Shared .d.ts, exported interfaces | Parse exported types |
| **Pydantic** | schemas/, models/ with BaseModel | Parse model definitions |
| **Zod** | schemas/ with z.object | Parse schema definitions |

**Contract Source Priority:**

1. Generated specs (openapi.json) - most accurate
2. Schema definitions (Pydantic, Zod) - source of truth
3. Type exports (TypeScript .d.ts) - consu

Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

Install

Pinned to the audited commit ea59a0c

git clone https://github.com/alinaqi/claude-bootstrap
git -C claude-bootstrap checkout ea59a0c53d79

The author documents no install command, so this uses the repository directly. Our verdict covers commit ea59a0c53d79 only.

Works in which agents?

What its own text says, not a test

AgentStatusMore
Claude CodeMentioned in its text, not testedSkills for Claude Code · About Claude Code
“Mentioned” means the skill’s text names the agent. We have not run it there.

Audit history

Every audit, every source change

  1. ✕BLOCK D 69 / 100

    first audit · source ea59a0c53d79

Questions about Workspace

What does the Workspace skill do?
What started as an opinionated Claude Code setup kit is now an autonomous AI engineering command center
Is Workspace safe to install?
No — not without reading the findings first. The audit graded it D (69/100) and found 1 critical or high issue in the source. Each one is listed on this page with the file and line it is on.
What can Workspace access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
Which assistants does Workspace work with?
Its documentation mentions claude-code. That is what the text claims, not a compatibility test we ran.
How current is this page?
The grade is for one exact copy of the source (ea59a0c53d79), read on 2026-10-14. The repository is watched, and a new audit runs when it changes — this is the first audit.

Sources

Spotted something wrong? Report a correction · Methodology

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.