nvidia-skillspector

Name: nvidia-skillspector
Author: jmxt3

32stars3forks

How to install nvidia-skillspector?

jmxt3/gitscape.ai/nvidia-skillspector

$npx -y skills add jmxt3/gitscape.ai --skill nvidia-skillspector

Installs into the current project.

Summary

Guides agents through working with the NVIDIA/SkillSpector codebase (Python, TypeScript, Shell). Use when extending, debugging, or navigating SkillSpector, or when the user mentions 'SkillSpector', 'NVIDIA/SkillSpector', or asks about its architecture, modules, or public API. Not

Files · 1

View on GitHub

SKILL.md

1	# Skillspector Code Skill
2
3	## Overview
4
5	Security scanner for AI agent skills. Detect vulnerabilities, malicious patterns, and security risks before installing agent skills.
6
7	Top-level areas: `.github/`, `contrib/`, `docs/`, `extensions/`, `src/`, `tests/`.
8
9	Primary languages: Python, TypeScript, Shell. The codebase contains 282 public symbols across 69 source files.
10
11	Key symbols:
12
13	- `is_language_compatible` `def is_language_compatible(rule_id: str, detected_language: str) -> bool` — Return ``True`` when rule_id is reliable for detected_language.
14	- `annotate_findings` `def annotate_findings( issues: list[dict[str, object]], detected_language: str, ) -> list[dict[str, object]]` — Add a ``language_compatible`` field to each issue dict.
15	- `ApiKey` `class ApiKey` — A single API key with concurrency and rate-limit metadata.
16	- `ApiKey.available` `def available(self) -> bool` — ``True`` when this key can accept at least one more caller.
17	- `ApiKeyPool` `class ApiKeyPool` — Thread-safe pool of API keys with per-key concurrency slots.
18	- `ApiKeyPool.__init__` `def __init__(self, keys: list[ApiKey]) -> None`
19	- `ApiKeyPool.acquire` `def acquire(self, timeout: float \| None = None) -> ApiKey` — Acquire a slot on the least-loaded available key.
20	- `ApiKeyPool.try_acquire` `def try_acquire(self) -> ApiKey \| None` — Non-blocking acquire — returns a key immediately or ``None``.
21	- `ApiKeyPool.release` `def release(self, key: ApiKey, , success: bool = True) -> None` — Release a slot on key* back to the pool.
22	- `ApiKeyPool.record_retry_success` `def record_retry_success(self) -> None` — Increment the retry-success counter for reporting.
23	- `ApiKeyPool.rate_limits_hit` `def rate_limits_hit(self) -> int` — Total number of 429 responses encountered across all keys.
24	- `ApiKeyPool.retry_successes` `def retry_successes(self) -> int` — Total number of successful retries after a key switch.
25	- `ApiKeyPool.keys_configured` `def keys_configured(self) -> int` — Total number of keys in the pool.
26	- `ApiKeyPool.total_capacity` `def total_capacity(self) -> int` — Sum of ``max_concurrent`` across all keys.
27	- `ApiKeyPool.active_requests` `def active_requests(self) -> int` — Total active requests across all keys.
28	- `ApiKeyPool.snapshot` `def snapshot(self) -> dict[str, object]` — Return a snapshot dict suitable for report metadata.
29	- `PooledChatModel` `class PooledChatModel` — LangChain-compatible chat model wrapper with transparent key switching.
30	- `PooledChatModel.__init__` `def __init__( self, pool: ApiKeyPool, *, max_tokens: int = 4096, timeout: float = 30.0, max_retries: int = _MAX_RATE_LIMIT_RETRIES, ) -> None`
31	- `PooledChatModel.invoke` `def invoke(self, prompt: str) -> object` — Synchronous invoke with automatic key switching on rate-limit.
32	- `PooledChatModel.ainvoke` `def ainvoke(self, prompt: str) -> object` — Async invoke with automatic key switching on rate-limit.
33
34	…and 262 more — see `references/api.md`.
35
36	## When to Use
37
38	- Understanding the architecture and module layout of SkillSpector
39	- Extending or modifying SkillSpector consistent with its existing patterns
40	- Debugging issues by tracing through SkillSpector's modules and dependencies
41	- Setting up, running, or configuring SkillSpector
42	- Calling functions, classes, or methods in SkillSpector's public API
43
44	When NOT to use: General Python, TypeScript, Shell questions, tutorials, or tasks unrelated to the SkillSpector codebase.
45
46	Related: For general Python, TypeScript, Shell guidance, use language-specific skills instead.
47
48	## Core Process
49
50	### Step 1: Understand the Architecture
51
52	Read the existing code in SkillSpector before making changes. Check `references/architecture.md` to understand the module layout, dependency graph, and internal import structure. The goal is to extend existing patterns, not invent new ones.
53
54	### Step 2: Locate Relevant Modules
55
56	Use `references/api.md` to find the public symbols, functions, and classes relevant to the task. Trace the call chain through SkillSpector's internal imports to understand how the pieces connect.
57
58	### Step 3: Make Changes Following Existing Patterns
59
60	Implement the change consistent with SkillSpector's established conventions: naming patterns, error handling style, module organization, and test structure. Consistency matters more than personal preference.
61
62	### Step 4: Verify the Change
63
64	Run the project's test suite and confirm all tests pass. If no tests exist for the changed behavior, write them first. Check that no regressions were introduced in adjacent modules.
65
66	## Common Rationalizations
67
68	\| Rationalization \| Reali

People who liked this also viewed

6 picks

Type

nvidia-skillspector

32stars3forks

View on GitHub

How to install nvidia-skillspector?

jmxt3/gitscape.ai/nvidia-skillspector

$npx -y skills add jmxt3/gitscape.ai --skill nvidia-skillspector

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Skillspector Code Skill
2
3	## Overview
4
5	Security scanner for AI agent skills. Detect vulnerabilities, malicious patterns, and security risks before installing agent skills.
6
7	Top-level areas: `.github/`, `contrib/`, `docs/`, `extensions/`, `src/`, `tests/`.
8
9	Primary languages: Python, TypeScript, Shell. The codebase contains 282 public symbols across 69 source files.
10
11	Key symbols:
12
13	- `is_language_compatible` `def is_language_compatible(rule_id: str, detected_language: str) -> bool` — Return ``True`` when rule_id is reliable for detected_language.
14	- `annotate_findings` `def annotate_findings( issues: list[dict[str, object]], detected_language: str, ) -> list[dict[str, object]]` — Add a ``language_compatible`` field to each issue dict.
15	- `ApiKey` `class ApiKey` — A single API key with concurrency and rate-limit metadata.
16	- `ApiKey.available` `def available(self) -> bool` — ``True`` when this key can accept at least one more caller.
17	- `ApiKeyPool` `class ApiKeyPool` — Thread-safe pool of API keys with per-key concurrency slots.
18	- `ApiKeyPool.__init__` `def __init__(self, keys: list[ApiKey]) -> None`
19	- `ApiKeyPool.acquire` `def acquire(self, timeout: float \| None = None) -> ApiKey` — Acquire a slot on the least-loaded available key.
20	- `ApiKeyPool.try_acquire` `def try_acquire(self) -> ApiKey \| None` — Non-blocking acquire — returns a key immediately or ``None``.
21	- `ApiKeyPool.release` `def release(self, key: ApiKey, , success: bool = True) -> None` — Release a slot on key* back to the pool.
22	- `ApiKeyPool.record_retry_success` `def record_retry_success(self) -> None` — Increment the retry-success counter for reporting.
23	- `ApiKeyPool.rate_limits_hit` `def rate_limits_hit(self) -> int` — Total number of 429 responses encountered across all keys.
24	- `ApiKeyPool.retry_successes` `def retry_successes(self) -> int` — Total number of successful retries after a key switch.
25	- `ApiKeyPool.keys_configured` `def keys_configured(self) -> int` — Total number of keys in the pool.
26	- `ApiKeyPool.total_capacity` `def total_capacity(self) -> int` — Sum of ``max_concurrent`` across all keys.
27	- `ApiKeyPool.active_requests` `def active_requests(self) -> int` — Total active requests across all keys.
28	- `ApiKeyPool.snapshot` `def snapshot(self) -> dict[str, object]` — Return a snapshot dict suitable for report metadata.
29	- `PooledChatModel` `class PooledChatModel` — LangChain-compatible chat model wrapper with transparent key switching.
30	- `PooledChatModel.__init__` `def __init__( self, pool: ApiKeyPool, *, max_tokens: int = 4096, timeout: float = 30.0, max_retries: int = _MAX_RATE_LIMIT_RETRIES, ) -> None`
31	- `PooledChatModel.invoke` `def invoke(self, prompt: str) -> object` — Synchronous invoke with automatic key switching on rate-limit.
32	- `PooledChatModel.ainvoke` `def ainvoke(self, prompt: str) -> object` — Async invoke with automatic key switching on rate-limit.
33
34	…and 262 more — see `references/api.md`.
35
36	## When to Use
37
38	- Understanding the architecture and module layout of SkillSpector
39	- Extending or modifying SkillSpector consistent with its existing patterns
40	- Debugging issues by tracing through SkillSpector's modules and dependencies
41	- Setting up, running, or configuring SkillSpector
42	- Calling functions, classes, or methods in SkillSpector's public API
43
44	When NOT to use: General Python, TypeScript, Shell questions, tutorials, or tasks unrelated to the SkillSpector codebase.
45
46	Related: For general Python, TypeScript, Shell guidance, use language-specific skills instead.
47
48	## Core Process
49
50	### Step 1: Understand the Architecture
51
52	Read the existing code in SkillSpector before making changes. Check `references/architecture.md` to understand the module layout, dependency graph, and internal import structure. The goal is to extend existing patterns, not invent new ones.
53
54	### Step 2: Locate Relevant Modules
55
56	Use `references/api.md` to find the public symbols, functions, and classes relevant to the task. Trace the call chain through SkillSpector's internal imports to understand how the pieces connect.
57
58	### Step 3: Make Changes Following Existing Patterns
59
60	Implement the change consistent with SkillSpector's established conventions: naming patterns, error handling style, module organization, and test structure. Consistency matters more than personal preference.
61
62	### Step 4: Verify the Change
63
64	Run the project's test suite and confirm all tests pass. If no tests exist for the changed behavior, write them first. Check that no regressions were introduced in adjacent modules.
65
66	## Common Rationalizations
67
68	\| Rationalization \| Reali

People who liked this also viewed

6 picks

Type