Skip to content
Back to skills

Mention Wire

ASecurity

Analyze codebase and inject @-mentions for traceability

  • 211 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 21, 2026
ai-agentstypescriptgobashsecuritydocumentation

Works with

  • claude code
  • cli

Security analysis

A100/100

Scanned September 21, 2026

npx -y skills add jmagly/ai-writing-guide --skill mention-wire --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Mention Wire?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Mention Wire
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/jmagly-mention-wire-d267d3aa/badge)](https://www.skillsdirectory.com/skills/jmagly-mention-wire-d267d3aa)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
namespace: aiwg
name: mention-wire
platforms: [all]
description: Analyze codebase and inject @-mentions for traceability
---

# @-Mention Wiring

Analyze codebase relationships and inject @-mentions for traceability.

## Research Foundation

- **REF-001**: BP-9 - Traceability from requirements to code to tests
- Claude Code 2.0.43: @-mention fixes for reliable nested loading

## Usage

```bash
/mention-wire                           # Analyze current directory
/mention-wire --dry-run                 # Show what would be added
/mention-wire --interactive             # Approve each mention
/mention-wire --auto                    # Apply high-confidence mentions
/mention-wire --confidence 90           # Set confidence threshold
```

## Options

| Option | Default | Description |
|--------|---------|-------------|
| --target | . | Directory to analyze |
| --dry-run | false | Show proposed changes without applying |
| --interactive | false | Prompt for approval per file |
| --auto | false | Apply mentions above confidence threshold |
| --confidence | 80 | Minimum confidence % for auto mode |

## Process

### 1. Scan Directory

Identify files and their types:
- Source code (`.ts`, `.js`, `.py`, `.go`, etc.)
- Test files (`*.test.*`, `*.spec.*`, `test_*`)
- SDLC artifacts (`.aiwg/**/*.md`)
- Documentation (`docs/**/*.md`)

### 2. Analyze Relationships

Detect relationships using heuristics:

| Pattern | Inferred @-mention | Confidence |
|---------|-------------------|------------|
| File in `src/auth/` | `@.aiwg/requirements/UC-*-auth*.md` | 85% |
| File named `*test*.ts` | `@$AIWG_ROOT/src/{corresponding-source}.ts` | 92% |
| Comment `// UC-001` | `@.aiwg/requirements/UC-001.md` | 95% |
| Comment `// ADR-005` | `@.aiwg/architecture/adrs/ADR-005*.md` | 90% |
| JSDoc `@implements` | Parse and validate | 98% |
| Import statement | `@{imported-file}` | 88% |

### 3. Generate Suggestions

Output format:
```
src/services/auth/login.ts (confidence: 85%)
  + @.aiwg/requirements/UC-003-user-auth.md (name match)
  + @.aiwg/architecture/adrs/ADR-005-jwt-strategy.md (comment: "JWT")

test/integration/auth.test.ts (confidence: 92%)
  + @$AIWG_ROOT/src/services/auth/login.ts (test-to-source)
  + @.aiwg/requirements/UC-003-user-auth.md (inherited from source)
```

### 4. Apply Changes

Depending on mode:
- `--dry-run`: Display only
- `--interactive`: Prompt per file
- `--auto`: Apply above threshold

## Placement Rules

### Code Files

Add @-mentions to file header:

```typescript
/**
 * @file Authentication Service
 * @implements @.aiwg/requirements/UC-003-user-auth.md
 * @architecture @.aiwg/architecture/adrs/ADR-005-jwt-strategy.md
 * @security @.aiwg/security/controls/authn-001.md
 * @tests @test/integration/auth.test.ts
 */
```

### Markdown Files

Add to References section:

```markdown
## References

- @.aiwg/requirements/user-stories.md - Functional requirements
- @.aiwg/architecture/software-architecture-doc.md - Architecture
```

## Examples

```bash
# Preview what would be wired
/mention-wire --dry-run

# Wire with interactive approval
/mention-wire --interactive

# Auto-wire high confidence (>80%)
/mention-wire --auto

# Auto-wire with higher threshold
/mention-wire --auto --confidence 90
```

## CLI Equivalent

```bash
aiwg wire-mentions [--target <dir>] [--dry-run] [--interactive] [--auto]
```

## Related Commands

- `/mention-validate` - Validate @-mentions resolve
- `/mention-lint` - Lint @-mention style
- `/mention-report` - Generate traceability report
- `/mention-conventions` - Display conventions

Wire @-mentions for: $ARGUMENTS

Attribution

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

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.