CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

nemo-mbridge-perf-hierarchical-context-parallel

Operational guide for enabling hierarchical context parallelism in Megatron-Bridge, including config knobs, code anchors, pitfalls, and verification.

Quality

77%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/nemo-mbridge-perf-hierarchical-context-parallel/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

82%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A strong, highly actionable body: exact config snippets, exact commands, explicit constraints, pitfalls with failure modes, and a verification section with concrete success criteria. The main gaps are minor — some trimmable quoted upstream code, no recovery loop after the "HCP is not active" failure signal, and code anchors that could be pushed to a reference file.

Suggestions

Add a fix-and-retry step after the "If you only see CONTEXT_PARALLEL_GROUP, HCP is not active" check (e.g., confirm hierarchical_context_parallel_sizes was passed through the recipe and that use_decentralized_pg is False, then relaunch) to close the workflow feedback loop.

Move the verbatim upstream docstring and initialization.py excerpts into a single reference file, keeping only the line-anchored paths in SKILL.md, to tighten the body and improve progressive disclosure.

Drop the duplicate statement of the product constraint (Pitfall 5 restates "prod(hierarchical_context_parallel_sizes) == context_parallel_size" from Required constraints) or merge them into one place.

DimensionReasoningScore

Conciseness

The body is dense and operational throughout — no basic-concept padding, every section (Enablement, Code Anchors, Pitfalls, Verification) earns its tokens. Minor trim opportunities exist, such as the 8-line upstream docstring quote in model_parallel_config.py and the slightly redundant restatement of the product constraint in both "Required constraints" and Pitfall 5, keeping it just short of fully lean.

4 / 5

Actionability

The skill gives copy-paste-ready guidance end to end: an exact Bridge override snippet (context_parallel_size=4, cp_comm_type="a2a+p2p", hierarchical_context_parallel_sizes=[2,2]), exact runnable commands (uv run pytest on a named test file; a fully specified 4-GPU torch.distributed.run invocation with recipe and overrides), and explicit success criteria.

5 / 5

Workflow Clarity

There is a coherent arc from enablement through constraints to pitfalls to a Verification section with explicit validation checkpoints (log check for HIERARCHICAL_CONTEXT_PARALLEL_GROUPS, "Training completes at least one step") and a failure signal ("If you only see CONTEXT_PARALLEL_GROUP, HCP is not active"). It stops short of level 5 because the failure signal has no corresponding fix-and-retry loop guiding recovery when HCP is not active.

4 / 5

Progressive Disclosure

Structure is good: short body, clear section headers, and well-signaled one-level-deep references up front ("For what hierarchical CP is, when to use it, and the decision tree..., see: @docs/... and @skills/.../card.yaml"). It does not reach 5 because the inlined upstream code-anchor excerpts (docstrings and initialization.py fragments) are detail that could live in a separate reference file, and there is no reference organization beyond the two intro links.

4 / 5

Total

17

/

20

Passed

Description

62%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is a crisp, specific, third-person statement of what the skill provides, but it answers only the "what" question. Trigger phrases and their synonyms exist only in the separate when_to_use field, so the description itself fails the "when should Claude use it" half of completeness.

Suggestions

Append an explicit "Use when..." clause to the description itself (e.g., "Use when scaling context parallelism beyond KV heads with cp_comm_type=a2a+p2p, or when investigating a CP-config change that caused OOM or a regression") so completeness is not dependent on the separate when_to_use field.

Fold natural trigger synonyms into the description — "hierarchical CP", "a2a+p2p", "multi-level CP", "CP beyond KV heads" — to lift trigger term coverage to comprehensive.

Name one or two concrete actions (e.g., "set hierarchical_context_parallel_sizes", "verify HIERARCHICAL_CONTEXT_PARALLEL_GROUPS in logs") to move specificity from deliverable-listing to concrete-action coverage.

DimensionReasoningScore

Specificity

The description names a concrete domain ("hierarchical context parallelism in Megatron-Bridge") and lists several specific deliverables: "config knobs, code anchors, pitfalls, and verification". It stops short of naming distinct concrete actions (e.g., set hierarchical_context_parallel_sizes, verify HCP groups in logs), so it does not reach the comprehensive level 5.

4 / 5

Completeness

The "what" is clear (an operational guide to enabling hierarchical CP with config, anchors, pitfalls, and verification), but the description field itself contains no "Use when..." clause or equivalent trigger guidance — the when-guidance lives in a separate when_to_use field — capping completeness at 3 per the rubric guideline.

3 / 5

Trigger Term Quality

Relevant keywords are present ("hierarchical context parallelism", "Megatron-Bridge") but common variations users would naturally say — "hierarchical CP", "CP", "a2a+p2p", "CP beyond KV heads" — appear only in the separate when_to_use field, not in the description itself, so synonym coverage is incomplete.

3 / 5

Distinctiveness Conflict Risk

"hierarchical context parallelism in Megatron-Bridge" is a precise niche with essentially no overlap risk against generic skills; a user or query mentioning these terms would unambiguously route here.

5 / 5

Total

15

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 1 missing, 1 deeper-than-1-level

Warning

Total

15

/

16

Passed

Repository

NVIDIA/skills

Path

skills/nemo-mbridge-perf-hierarchical-context-parallel/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.