langsmith-cli 0.3.4__tar.gz → 0.4.0__tar.gz

{langsmith_cli-0.3.4 → langsmith_cli-0.4.0}/.claude-plugin/marketplace.json

@@ -6,14 +6,14 @@
   },
   "metadata": {
     "description": "LangSmith CLI plugin marketplace",
-    "version": "0.2.9"
+    "version": "0.4.0"
   },
   "plugins": [
     {
       "name": "langsmith-cli",
       "source": "./",
       "description": "A context-efficient interface for LangSmith observability and evaluations.",
-      "version": "0.2.9",
+      "version": "0.4.0",
       "author": {
         "name": "Gigaverse",
         "email": "aviadr1@gmail.com"

{langsmith_cli-0.3.4 → langsmith_cli-0.4.0}/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "langsmith-cli",
-  "version": "0.3.4",
+  "version": "0.4.0",
   "description": "A context-efficient interface for LangSmith observability and evaluations.",
   "author": {
     "name": "Aviad Rozenhek",

{langsmith_cli-0.3.4 → langsmith_cli-0.4.0}/CLAUDE.md

@@ -107,7 +107,10 @@ main.py (entry point)
 │   └── login
 ├── projects (group)
 │   ├── list
-│   └── create
+│   ├── get
+│   ├── create
+│   ├── update
+│   └── delete
 ├── runs (group)
 │   ├── list
 │   ├── get
@@ -122,20 +125,29 @@ main.py (entry point)
 │   ├── metadata-keys
 │   ├── fields
 │   ├── describe
-│   └── view-file
+│   ├── view-file
+│   └── export
 ├── datasets (group)
 │   ├── list
 │   ├── get
 │   ├── create
+│   ├── delete
 │   └── push
 ├── examples (group)
 │   ├── list
 │   ├── get
-│   └── create
+│   ├── create
+│   ├── update
+│   ├── delete
+│   └── from-run
 ├── prompts (group)
 │   ├── list
 │   ├── get
-│   └── push
+│   ├── pull
+│   ├── push
+│   ├── create
+│   ├── delete
+│   └── commits
 └── self (group)
     ├── detect
     └── update
@@ -143,18 +155,27 @@ main.py (entry point)
 
 ### Key Design Patterns
 
-**1. Lazy Loading Performance Pattern**
+**1. Direct Imports for Strong Types**
 ```python
-# ❌ BAD - Top-level import (loads SDK immediately)
-from langsmith import Client
+# ✅ GOOD - Direct imports of langsmith types for strong typing
+import langsmith
+from langsmith.schemas import Dataset, TracerSessionResult, Run
 
-# ✅ GOOD - Import inside command function
-@runs.command("list")
-def list_runs(...):
-    import langsmith  # Only loads when command executes
-    client = langsmith.Client()
+def resolve_dataset(client: langsmith.Client, name: str) -> Dataset:
+    ...
+
+# ❌ BAD - using `object` or `Any` as type (loses all type safety)
+def resolve_dataset(client: object, name: str) -> object: ...
+
+# ❌ BAD - TYPE_CHECKING / __future__ annotations workaround (unnecessary complexity)
+from __future__ import annotations
+from typing import TYPE_CHECKING
+if TYPE_CHECKING:
+    import langsmith  # Don't do this
 ```
 
+**Note on CLI startup performance:** `utils.py` already imports `langsmith` at module level (needed for shared helpers), and every command module imports `utils.py`. Since the cost is already paid at startup, import langsmith types directly everywhere — no lazy loading workarounds needed for type annotations. If a module does NOT import utils.py and doesn't need langsmith at import time, keep heavy imports inside command functions.
+
 **2. Dual Output Pattern (JSON vs Rich Tables)**
 ```python
 # All commands check ctx.obj.get("json") flag
@@ -530,6 +551,107 @@ def create_dataset(name="test", example_count=10) -> Dataset:
 - E2E tests validate full SDK interaction (require API key)
 - Aim for 100% coverage of new code
 
+### Handling Mixed stdout/stderr in Tests (CliRunner)
+
+Click's `CliRunner` mixes stdout and stderr. When commands use `logger.info()` before JSON output, the `--json` test output may contain non-JSON lines. Use the shared `parse_json_output()` helper from conftest:
+
+```python
+from conftest import parse_json_output
+
+def test_json_output(runner, tmp_path):
+    result = runner.invoke(cli, ["--json", "runs", "export", str(tmp_path), ...])
+    data = parse_json_output(result.output)  # Finds last JSON line
+    assert data["status"] == "success"
+```
+
+**Anti-pattern:** Don't use `json.loads(result.output)` directly when commands emit progress messages — it will fail with `JSONDecodeError`.
+
+### SDK Exception Types in Tests
+
+Always use real SDK exception types in test side_effects, never bare `Exception`:
+
+```python
+# ❌ BAD - bare Exception won't be caught by specific handlers
+mock_client.delete_example.side_effect = Exception("Not found")
+
+# ✅ GOOD - matches the actual exception hierarchy
+from langsmith.utils import LangSmithNotFoundError
+mock_client.delete_example.side_effect = LangSmithNotFoundError("Not found")
+```
+
+## Reusable Patterns Across Commands
+
+### Name-or-ID Resolution (`resolve_project()`, `resolve_dataset()`)
+
+When a command accepts a name or ID argument, use the existing resolve helpers:
+- `projects.py: resolve_project(client, name_or_id, include_stats=True)`
+- `datasets.py: resolve_dataset(client, name_or_id)`
+
+These helpers follow a consistent pattern:
+1. Auto-detect UUIDs via `_looks_like_uuid()` to save an API call
+2. Fall back from name to ID lookup
+3. Catch `(LangSmithNotFoundError, LangSmithError, ValueError)` for friendly errors
+4. Raise `click.ClickException` on not-found
+
+```python
+from langsmith_cli.commands.projects import resolve_project
+
+project = resolve_project(client, name_or_id, include_stats=True)
+```
+
+### Shared Filter Building (utils.py: `build_runs_list_filter()`)
+
+Never manually reconstruct status/tag/time filters. Use the canonical helper:
+
+```python
+combined_filter, error_filter = build_runs_list_filter(
+    filter_=filter_, status=status, tag=tag, since=since, last=last
+)
+```
+
+### Shared Multi-Project Fetching (utils.py: `fetch_from_projects()`)
+
+Never write manual project iteration loops. Use `fetch_from_projects()`:
+
+```python
+def _fetch_runs(c, proj, **kw):
+    if proj is not None:
+        return c.list_runs(project_name=proj, **kw)
+    return c.list_runs(**kw)
+
+result = fetch_from_projects(client, projects, _fetch_runs, project_query=pq, ...)
+raise_if_all_failed_with_suggestions(result, client, pq, logger, "runs")
+```
+
+### Destructive Operations Pattern
+
+For delete/destructive commands, follow this pattern:
+1. Require `--confirm` flag (skip with `click.confirm(abort=True)`)
+2. Resolve entity first (verify it exists), then perform the operation
+3. Use specific SDK exceptions (`LangSmithNotFoundError`), never broad `except Exception`
+4. In JSON mode, return `{"status": "success", ...}` on success
+
+### Filename Sanitization
+
+When writing user-controlled filenames, sanitize with regex:
+```python
+import re
+safe = re.sub(r'[<>:"/\\|?*\x00-\x1f]', "_", name)
+safe = safe.strip(". ")
+```
+
+### Common Anti-Patterns to Avoid
+
+1. **`hasattr()` on Pydantic models** → Use `p.field is not None` instead
+2. **Broad `except Exception`** in loops → Catch specific SDK exceptions
+3. **Manual filter construction** → Use `build_runs_list_filter()`
+4. **Manual project iteration** → Use `fetch_from_projects()`
+5. **Duplicated utility logic** → Extract to helpers (e.g., `normalize_split()`)
+6. **`--flag default=True is_flag=True`** → No-op; use `--flag/--no-flag` pair
+7. **ThreadPoolExecutor for local file I/O** → Sequential is faster (no GIL benefit)
+8. **`TYPE_CHECKING` / `__future__` annotations workaround** → Import types directly; langsmith is already loaded at startup
+9. **`client: object` or `client: Any`** → Use `client: langsmith.Client` for strong types
+
 ## Engineering Standards (from docs/AGENTS.md)
 
 ### 1. Type Safety (Zero Tolerance for Weak Types)
@@ -542,10 +664,13 @@ def create_dataset(name="test", example_count=10) -> Dataset:
 - ✅ Never create duplicate response models - reuse SDK models with field selection
 - **Current Status**: Fully implemented - all commands use SDK Pydantic models with type-safe attribute access
 
-### 2. Performance (100ms Rule)
-- ❌ NO top-level imports of heavy libraries (langsmith, rich, pandas)
-- ✅ Libraries imported inside command functions (lazy loading)
-- **Current Status**: Properly implemented
+### 2. Performance (Keep CLI Fast)
+- ❌ Don't add new top-level imports of heavy libraries to modules that don't already need them
+- ❌ Don't introduce new heavy dependencies without measuring CLI startup impact
+- ✅ `langsmith` and `rich` are already loaded at startup (via utils.py) — use them directly for strong types
+- ✅ If adding a new module that doesn't need langsmith at import time, keep heavy imports inside functions
+- ✅ Always prefer strong types over lazy loading workarounds — don't sacrifice type safety for startup time
+- **Current Status**: langsmith/rich load at startup (~300ms). Acceptable. Don't make it worse.
 
 ### 3. Architecture (Pure Logic vs View)
 - Logic Layer: Returns Pydantic models/typed objects, never prints

{langsmith_cli-0.3.4 → langsmith_cli-0.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: langsmith-cli
-Version: 0.3.4
+Version: 0.4.0
 Summary: Context-efficient CLI for LangSmith. Built for humans and agents.
 Project-URL: Homepage, https://github.com/aviadr1/langsmith-cli
 Project-URL: Repository, https://github.com/aviadr1/langsmith-cli

{langsmith_cli-0.3.4 → langsmith_cli-0.4.0}/pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 name = "langsmith-cli"
 # IMPORTANT: When bumping this version, also update .claude-plugin/plugin.json
-version = "0.3.4"
+version = "0.4.0"
 description = "Context-efficient CLI for LangSmith. Built for humans and agents."
 readme = "README.md"
 requires-python = ">=3.12"

{langsmith_cli-0.3.4 → langsmith_cli-0.4.0}/skills/langsmith/SKILL.md

@@ -103,7 +103,12 @@ langsmith-cli --json runs list --project my-project --limit 5 2>&1
 - `langsmith-cli --json projects list [OPTIONS]`: List all projects.
   - `--fields <comma-separated>`: Select specific fields (e.g., `id,name`)
   - `--output <file>`: Write to file instead of stdout
+- `langsmith-cli --json projects get <name-or-id>`: Get project details (UUID auto-detected).
+  - `--include-stats/--no-stats`: Include/exclude run statistics (default: include)
+  - `--fields <comma-separated>`: Select fields
 - `langsmith-cli --json projects create <name>`: Create a new project.
+- `langsmith-cli --json projects update <name-or-id> --name <new> --description <desc>`: Update project.
+- `langsmith-cli --json projects delete <name-or-id> --confirm`: Delete a project.
 
 ### Runs (Traces)
 - `langsmith-cli --json runs list [OPTIONS]`: List recent runs.
@@ -192,6 +197,17 @@ langsmith-cli --json runs list --project my-project --limit 5 2>&1
   - Returns: `{"fields": [{"path": "inputs.query", "type": "string", "present_pct": 98.0, "length": {"min": 5, "max": 500, "avg": 89}, "languages": {"en": 80.0}}, ...], "total_runs": 100}`
   - Example: `langsmith-cli --json runs describe --include inputs,outputs`
   - Example: `langsmith-cli --json runs describe --project my-project --no-language`
+- `langsmith-cli runs export <directory> [OPTIONS]`: Export runs as individual JSON files.
+  - `--project <name>`: Project name (required)
+  - `--limit <n>`: Max runs to export (default: 50)
+  - `--status <success|error>`: Filter by status
+  - `--roots`: Export only root traces
+  - `--tag <tag>`: Filter by tag (repeatable)
+  - `--last <duration>`: Time window (e.g., `24h`, `7d`)
+  - `--fields <comma-separated>`: Reduce exported file size
+  - `--filename-pattern <pattern>`: Custom filenames (placeholders: `{run_id}`, `{name}`, `{index}`, `{trace_id}`)
+  - Example: `langsmith-cli runs export ./traces --project my-project --roots --limit 100`
+  - Example: `langsmith-cli --json runs export ./errors --project my-project --status error --last 24h --fields name,inputs,outputs,error`
 
 ### Datasets & Examples
 - `langsmith-cli --json datasets list [OPTIONS]`: List datasets.
@@ -199,18 +215,29 @@ langsmith-cli --json runs list --project my-project --limit 5 2>&1
   - `--output <file>`: Write to file instead of stdout
 - `langsmith-cli --json datasets get <id> [--fields id,name,description]`: Get dataset details.
 - `langsmith-cli --json datasets create <name>`: Create a dataset.
+- `langsmith-cli --json datasets delete <name-or-id> --confirm`: Delete a dataset.
+- `langsmith-cli --json datasets push <file.jsonl> --dataset <name>`: Upload examples from JSONL.
 - `langsmith-cli --json examples list --dataset <name> [OPTIONS]`: List examples in a dataset.
   - `--fields <comma-separated>`: Select fields (e.g., `id,inputs,outputs`)
   - `--output <file>`: Write to file instead of stdout
 - `langsmith-cli --json examples get <id> [--fields id,inputs,outputs]`: Get example details.
 - `langsmith-cli --json examples create --dataset <name> --inputs <json> --outputs <json>`: Add an example.
+- `langsmith-cli --json examples update <id> --inputs <json> --outputs <json>`: Update an example.
+- `langsmith-cli --json examples delete <id> [<id>...] --confirm`: Delete examples (supports bulk).
+- `langsmith-cli --json examples from-run <run-id> --dataset <name>`: Create example from a run.
 
 ### Prompts
 - `langsmith-cli --json prompts list [OPTIONS]`: List prompt repositories.
   - `--fields <comma-separated>`: Select fields (e.g., `repo_handle,description`)
   - `--output <file>`: Write to file instead of stdout
 - `langsmith-cli --json prompts get <name> [--commit <hash>]`: Fetch a prompt template.
+- `langsmith-cli --json prompts pull <name> [--commit <hash>]`: Pull full prompt content (manifest, examples).
+  - `--include-model`: Include model configuration
+  - `--fields <comma-separated>`: Select fields
 - `langsmith-cli --json prompts push <name> <file_path>`: Push a local file as a prompt.
+- `langsmith-cli --json prompts create <name> [--description <text>]`: Create a new prompt.
+- `langsmith-cli --json prompts delete <name> --confirm`: Delete a prompt.
+- `langsmith-cli --json prompts commits <name> [--limit N]`: List prompt versions.
 
 ### Self (Installation Management)
 - `langsmith-cli self detect`: Show installation details (version, install method, paths).

{langsmith_cli-0.3.4 → langsmith_cli-0.4.0}/skills/langsmith/references/datasets.md

@@ -9,12 +9,16 @@ langsmith-cli --json datasets list [OPTIONS]
 ```
 
 **Options:**
-- `--limit INTEGER` - Maximum results (default: 10)
+- `--limit INTEGER` - Maximum results (default: 20)
 - `--name TEXT` - Filter by exact dataset name
-- `--name-contains TEXT` - Filter by name substring (case-insensitive)
+- `--name-contains TEXT` - Filter by name substring
 - `--dataset-ids TEXT` - Comma-separated list of dataset UUIDs
 - `--data-type TEXT` - Filter by type: `kv`, `llm`, or `chat`
-- `--metadata JSON` - Filter by metadata (JSON object)
+- `--metadata TEXT` - Filter by metadata (JSON string)
+- `--exclude TEXT` - Exclude items containing substring (repeatable)
+- `--fields TEXT` - Comma-separated field names to include
+- `--count` - Output only the count of results
+- `--output TEXT` - Write output to file (JSONL format)
 
 **Output Fields:**
 - `id` (UUID) - Dataset identifier
@@ -43,17 +47,21 @@ langsmith-cli --json datasets list --data-type llm
 Get dataset details.
 
 ```bash
-langsmith-cli --json datasets get <dataset-id-or-name>
+langsmith-cli --json datasets get <dataset-id> [OPTIONS]
 ```
 
 **Arguments:**
-- `dataset-id-or-name` (required) - Dataset UUID or exact name
+- `dataset-id` (required) - Dataset UUID
+
+**Options:**
+- `--fields TEXT` - Comma-separated field names to include
+- `--output TEXT` - Write output to file (JSON format)
 
 **Output:** Complete dataset object with all metadata
 
 **Example:**
 ```bash
-langsmith-cli --json datasets get "my-test-dataset"
+langsmith-cli --json datasets get "ae99b6fa-a6db-4f1c-8868-bc6764f4c29e"
 ```
 
 ### `datasets create`
@@ -69,8 +77,7 @@ langsmith-cli --json datasets create <name> [OPTIONS]
 
 **Options:**
 - `--description TEXT` - Dataset description
-- `--type TEXT` - Dataset type: `kv` (default), `llm`, or `chat`
-- `--metadata JSON` - Custom metadata (JSON object)
+- `--type [kv|llm|chat]` - Dataset type (default: kv)
 
 **Output:** Created dataset object
 
@@ -78,8 +85,28 @@ langsmith-cli --json datasets create <name> [OPTIONS]
 ```bash
 langsmith-cli --json datasets create "qa-pairs" \
   --description "Question answering test set" \
-  --type kv \
-  --metadata '{"source": "production", "version": "1.0"}'
+  --type kv
+```
+
+### `datasets delete`
+
+Delete a dataset by name or ID.
+
+```bash
+langsmith-cli --json datasets delete <name-or-id> --confirm
+```
+
+**Arguments:**
+- `name-or-id` (required) - Dataset name or UUID (auto-detected)
+
+**Options:**
+- `--confirm` - Skip confirmation prompt (required for non-interactive use)
+
+**Output:** `{"status": "success", "name": "<dataset-name>"}`
+
+**Example:**
+```bash
+langsmith-cli --json datasets delete "old-test-data" --confirm
 ```
 
 ### `datasets push`
@@ -87,27 +114,24 @@ langsmith-cli --json datasets create "qa-pairs" \
 Bulk upload examples from JSONL file.
 
 ```bash
-langsmith-cli --json datasets push <dataset-name> <file.jsonl> [OPTIONS]
+langsmith-cli --json datasets push <file.jsonl> [OPTIONS]
 ```
 
 **Arguments:**
-- `dataset-name` (required) - Target dataset name (creates if doesn't exist)
 - `file.jsonl` (required) - Path to JSONL file
 
+**Options:**
+- `--dataset TEXT` - Target dataset name (creates if doesn't exist; defaults to filename)
+
 **JSONL Format:**
 ```jsonl
 {"inputs": {"query": "What is AI?"}, "outputs": {"answer": "Artificial Intelligence..."}}
 {"inputs": {"query": "Define ML"}, "outputs": {"answer": "Machine Learning..."}}
 ```
 
-**Options:**
-- `--description TEXT` - Description for new dataset
-- `--type TEXT` - Type for new dataset: `kv`, `llm`, or `chat`
-
 **Output:** Upload summary with count of examples added
 
 **Example:**
 ```bash
-langsmith-cli --json datasets push "my-dataset" examples.jsonl
+langsmith-cli --json datasets push examples.jsonl --dataset "my-dataset"
 ```
-

{langsmith_cli-0.3.4 → langsmith_cli-0.4.0}/skills/langsmith/references/examples.md

@@ -10,7 +10,7 @@ langsmith-cli --json examples list [OPTIONS]
 
 **Options:**
 - `--dataset TEXT` (required) - Dataset name or UUID
-- `--limit INTEGER` - Maximum results (default: 10, max: 100)
+- `--limit INTEGER` - Maximum results (default: 20)
 - `--offset INTEGER` - Skip N examples (default: 0)
 - `--example-ids TEXT` - Comma-separated list of example UUIDs
 - `--filter TEXT` - Advanced FQL query
@@ -19,6 +19,10 @@ langsmith-cli --json examples list [OPTIONS]
 - `--as-of TEXT` - Version tag or ISO timestamp
 - `--inline-s3-urls BOOLEAN` - Inline S3 URLs: `true` or `false`
 - `--include-attachments BOOLEAN` - Include attachments: `true` or `false`
+- `--exclude TEXT` - Exclude items containing substring (repeatable)
+- `--fields TEXT` - Comma-separated field names to include
+- `--count` - Output only the count of results
+- `--output TEXT` - Write output to file (JSONL format)
 
 **Output Fields:**
 - `id` (UUID) - Example identifier
@@ -61,6 +65,8 @@ langsmith-cli --json examples get <example-id> [OPTIONS]
 
 **Options:**
 - `--as-of TEXT` - Version tag or ISO timestamp
+- `--fields TEXT` - Comma-separated field names to include
+- `--output TEXT` - Write output to file (JSON format)
 
 **Output:** Complete example object
 
@@ -83,7 +89,6 @@ langsmith-cli --json examples create [OPTIONS]
 - `--outputs JSON` - Expected output data as JSON object
 - `--metadata JSON` - Custom metadata as JSON object
 - `--split TEXT` - Split name (e.g., "train", "test", "validation")
-- `--source-run-id UUID` - Source run UUID if example from trace
 
 **Output:** Created example object
 
@@ -104,3 +109,78 @@ langsmith-cli --json examples create \
   --split train
 ```
 
+### `examples update`
+
+Update an existing example's inputs, outputs, metadata, or split.
+
+```bash
+langsmith-cli --json examples update <example-id> [OPTIONS]
+```
+
+**Arguments:**
+- `example-id` (required) - Example UUID
+
+**Options:**
+- `--inputs JSON` - New input data as JSON
+- `--outputs JSON` - New output data as JSON
+- `--metadata JSON` - New metadata as JSON
+- `--split TEXT` - New split name
+
+At least one option is required.
+
+**Output:** Updated example data
+
+**Example:**
+```bash
+langsmith-cli --json examples update <uuid> \
+  --outputs '{"answer": "Updated answer"}' \
+  --metadata '{"reviewed": true}'
+```
+
+### `examples delete`
+
+Delete one or more examples by ID. Supports bulk deletion with partial failure reporting.
+
+```bash
+langsmith-cli --json examples delete <example-id> [<example-id>...] [OPTIONS]
+```
+
+**Arguments:**
+- `example-ids` (required) - One or more example UUIDs
+
+**Options:**
+- `--confirm` - Skip confirmation prompt
+
+**Output:** `{"status": "success", "deleted": [...], "errors": [...]}`
+
+**Examples:**
+```bash
+# Delete single example
+langsmith-cli --json examples delete <uuid> --confirm
+
+# Bulk delete
+langsmith-cli --json examples delete <uuid1> <uuid2> <uuid3> --confirm
+```
+
+### `examples from-run`
+
+Create an example from a run's inputs and outputs.
+
+```bash
+langsmith-cli --json examples from-run <run-id> --dataset <name>
+```
+
+**Arguments:**
+- `run-id` (required) - Run UUID to create example from
+
+**Options:**
+- `--dataset TEXT` (required) - Dataset name to add the example to
+
+**Output:** Created example object
+
+**Example:**
+```bash
+# Turn a good run into a training example
+langsmith-cli --json examples from-run <run-uuid> --dataset "training-data"
+```
+

langsmith_cli-0.4.0/skills/langsmith/references/projects.md

@@ -0,0 +1,141 @@
+## Projects
+
+### `projects list`
+
+List all LangSmith projects (sessions).
+
+```bash
+langsmith-cli --json projects list [OPTIONS]
+```
+
+**Options:**
+- `--limit INTEGER` - Maximum number of projects to return (default: 100, use 0 for no limit)
+- `--name TEXT` - Filter by exact project name
+- `--name-pattern TEXT` - Filter by name with wildcards (e.g. `'*prod*'`)
+- `--name-regex TEXT` - Filter by name with regex
+- `--reference-dataset-name TEXT` - Filter projects by reference dataset name
+- `--has-runs` - Show only projects with runs (run_count > 0)
+- `--sort-by TEXT` - Sort by field (name, run_count). Prefix with `-` for descending
+- `--exclude TEXT` - Exclude items containing substring (repeatable)
+- `--fields TEXT` - Comma-separated field names to include
+- `--count` - Output only the count of results
+- `--output TEXT` - Write output to file (JSONL format)
+- `--format [table|json|csv|yaml]` - Output format
+
+**Output Fields:**
+- `id` (UUID) - Project identifier
+- `name` (string) - Project name
+- `description` (string|null) - Project description
+- `created_at` (datetime) - Creation timestamp
+- `run_count` (integer) - Number of runs in project
+- `latency_p50` (float|null) - Median latency in seconds
+- `latency_p99` (float|null) - 99th percentile latency
+- `first_start_time` (datetime|null) - First run start time
+- `last_start_time` (datetime|null) - Most recent run start time
+- `feedback_stats` (object|null) - Feedback statistics
+- `total_tokens` (integer|null) - Total tokens used
+- `prompt_tokens` (integer|null) - Prompt tokens
+- `completion_tokens` (integer|null) - Completion tokens
+- `total_cost` (float|null) - Total cost in USD
+
+**Example:**
+```bash
+langsmith-cli --json projects list --limit 5
+```
+
+### `projects create`
+
+Create a new project.
+
+```bash
+langsmith-cli --json projects create <name> [OPTIONS]
+```
+
+**Arguments:**
+- `name` (required) - Project name
+
+**Options:**
+- `--description TEXT` - Project description
+
+**Output:** Created project object
+
+**Example:**
+```bash
+langsmith-cli --json projects create "my-experiment" --description "Testing new prompt"
+```
+
+### `projects get`
+
+Get details of a single project by name or ID.
+
+```bash
+langsmith-cli --json projects get <name-or-id> [OPTIONS]
+```
+
+**Arguments:**
+- `name-or-id` (required) - Project name or UUID (UUIDs auto-detected)
+
+**Options:**
+- `--include-stats/--no-stats` - Include/exclude run statistics (default: include)
+- `--fields TEXT` - Comma-separated fields to return
+- `--output FILE` - Write output to file
+
+**Output:** Full project object (same fields as `projects list`)
+
+**Examples:**
+```bash
+# Get by name
+langsmith-cli --json projects get "my-project"
+
+# Get by UUID (auto-detected, saves an API call)
+langsmith-cli --json projects get "f47ac10b-58cc-4372-a567-0e02b2c3d479"
+
+# Get with field pruning
+langsmith-cli --json projects get "my-project" --fields name,run_count,error_rate
+```
+
+### `projects update`
+
+Update a project's name or description.
+
+```bash
+langsmith-cli --json projects update <name-or-id> [OPTIONS]
+```
+
+**Arguments:**
+- `name-or-id` (required) - Project name or UUID
+
+**Options:**
+- `--name TEXT` - New project name
+- `--description TEXT` - New project description
+
+At least one of `--name` or `--description` is required.
+
+**Output:** Updated project object
+
+**Example:**
+```bash
+langsmith-cli --json projects update "old-name" --name "new-name" --description "Updated"
+```
+
+### `projects delete`
+
+Delete a project.
+
+```bash
+langsmith-cli --json projects delete <name-or-id> [OPTIONS]
+```
+
+**Arguments:**
+- `name-or-id` (required) - Project name or UUID
+
+**Options:**
+- `--confirm` - Skip confirmation prompt
+
+**Output:** Success status
+
+**Example:**
+```bash
+langsmith-cli --json projects delete "test-project" --confirm
+```
+