pkg/embedded/templates/specledger/skills/sl-comment/SKILL.md
# sl-comment Skill **When to Load**: Triggered when AI commands mention "review comments", "comment", or reference `sl comment` CLI. ## Overview The `sl comment` CLI provides review comment management for SpecLedger projects. Comments are stored in Supabase and accessed via the PostgREST API. ## Subcommands | Command | Purpose | Output Mode | |---------|---------|-------------| | `sl comment list [branch-name] [-R owner/repo]` | List all comments (compact) | Truncated previews, reply counts
npx skillsauth add specledger/specledger pkg/embedded/templates/specledger/skills/sl-commentInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
When to Load: Triggered when AI commands mention "review comments", "comment", or reference sl comment CLI.
The sl comment CLI provides review comment management for SpecLedger projects. Comments are stored in Supabase and accessed via the PostgREST API.
| Command | Purpose | Output Mode |
|---------|---------|-------------|
| sl comment list [branch-name] [-R owner/repo] | List all comments (compact) | Truncated previews, reply counts |
| sl comment show <id> | Full comment details | Complete content, all replies |
| sl comment reply <id> "msg" | Reply to a comment | Minimal confirmation |
| sl comment resolve <id> --reason "text" | Mark comment resolved (reason required, posted as reply) | Minimal confirmation |
Use sl comment list when:
Use sl comment show when:
Reply (sl comment reply):
Resolve (sl comment resolve --reason):
--reason is required and auto-posts a replysl comment list --status open --json
[
{
"id": "abc123",
"file_path": "spec.md",
"line": 42,
"content_preview": "This requirement is ambiguous...",
"author": "alice",
"reply_count": 2
}
]
Note: content_preview is truncated to 120 characters. Use sl comment show for full content.
sl comment show abc123 --json
{
"id": "abc123",
"file_path": "spec.md",
"line": 42,
"content": "This requirement is ambiguous - does it apply to all users or just admins?",
"selected_text": "all users",
"author": "alice",
"status": "open",
"created_at": "2026-03-05T10:30:00Z",
"replies": [
{
"id": "reply1",
"author": "bob",
"content": "Good point - we should clarify this in the spec.",
"created_at": "2026-03-05T11:00:00Z"
}
]
}
# 1. List all open comments
sl comment list --status open --json
# 2. For each comment, get full details
sl comment show <id> --json
# 3. After addressing in code, resolve with reason (auto-posts reply)
sl comment resolve <id> --reason "Fixed in commit abc123. Added role check."
When automatic repo detection fails (e.g., non-standard remote URL format, missing origin remote,
or non owner/repo path structure), use --repo / -R to specify the owner/repo manually:
sl comment list --repo owner/repo --status open --json
The sl revise command also supports --repo / -R for the same purpose.
# List all open comments, then resolve multiple
sl comment list --status open --json
# ... after addressing all ...
sl comment resolve id1 id2 id3 --reason "Batch resolved: all addressed in latest revision"
| Error | Cause | Solution |
|-------|-------|----------|
| Exit code 1 (silent) | No auth token | Run sl auth login |
| "cannot parse owner/repo from remote URL" | Non-standard remote URL format or missing origin remote | Use --repo owner/repo to specify manually |
| "comment not found" | Invalid ID | Verify ID from list output |
| Network error | Supabase unavailable | Retry or check connectivity |
The CLI follows the token-efficient output pattern:
Use --json flag for programmatic parsing by AI agents.
testing
Manage agent skills from the skills.sh registry — search, install, audit, and remove community-built skills
testing
# sl Issue Tracking ## Overview `sl issue` is the built-in issue tracker for SpecLedger. Use it for multi-session work with complex dependencies; use TodoWrite for simple single-session tasks. ## When to Use sl issue vs TodoWrite ### Use sl issue when: - **Multi-session work** - Tasks spanning multiple compaction cycles or days - **Complex dependencies** - Work with blockers, prerequisites, or hierarchical structure - **Knowledge work** - Strategic documents, research, or tasks with fuzzy bo
testing
# sl Dependency Management **When to Load**: Triggered when tasks involve cross-repo specification dependencies, `sl deps` commands, artifact caching, or multi-repo dependency resolution. ## Overview `sl deps` manages external specification dependencies between repositories. Dependencies are declared in `specledger.yaml`, cached locally at `~/.specledger/cache/`, and optionally symlinked into your project's artifacts directory for direct access. ## Subcommands | Command | Purpose | Output M
development
# sl Audit Skill ## Overview Codebase reconnaissance skill for understanding project structure, tech stack, and module organization. This skill provides patterns for efficient codebase exploration without AI orchestration overhead. **Layer**: L3 (Skill) - Passive context injection **Use when**: You need to understand an unfamiliar codebase or validate architecture assumptions ## When to Load Load this skill when: - First encounter with an unfamiliar codebase - Need to understand project str