dspy

DSPy: declarative LM programs, auto-optimize prompts, RAG.

Quality

57%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./optional-skills/mlops/research/dspy/SKILL.md

Quality

Content

65%

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

The body is highly actionable with extensive executable code and a sensible topical ordering, but it is verbose for a SKILL.md overview and largely duplicates content that already lives in the provided reference files rather than deferring to them. Progressive disclosure and conciseness are the main drag on an otherwise strong, code-rich skill.

Suggestions

Trim the inline Modules, Optimizers, and Common Patterns sections to concise overviews and link to references/modules.md, optimizers.md, and examples.md at each relevant heading to improve both conciseness and progressive disclosure.

Add explicit validation/feedback steps where workflows carry risk (e.g., after optimizer.compile, show how to run dspy.evaluate.Evaluate and only proceed if the metric improves) to lift workflow_clarity toward 3.

Move the inline LM-provider configuration and Comparison-to-other-approaches material into a reference file, keeping SKILL.md as a lean overview with quick-start plus well-signaled one-level-deep links.

Dimension	Reasoning	Score
Conciseness	The ~580-line body is a full tutorial with mild conceptual prose ("Signatures define the structure of your AI task...") and large volumes of inline module/optimizer/pattern detail; the concrete code earns its place but the overall token budget is high and could be tightened, matching anchor 2.	2 / 3
Actionability	It provides abundant executable, copy-paste-ready code with imports, configuration, and usage (e.g., "pip install dspy", full dspy.Signature/dspy.Module classes), matching the fully-executable anchor 3.	3 / 3
Workflow Clarity	A logical install→quickstart→concepts→best-practices ordering and a "Start Simple, Iterate" sequence exist, but there are no explicit validation checkpoints or feedback loops, fitting anchor 2 rather than the validated-workflow bar of 3.	2 / 3
Progressive Disclosure	A "See Also" section signals references/modules.md, optimizers.md, and examples.md (all verified to exist, one level deep), but the module/optimizer/example content is duplicated inline rather than appropriately split, matching the "content that should be separate is inline" anchor 2.	2 / 3
	Total	9 / 12 Passed

Description

50%

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 concisely names DSPy and three core capabilities but relies on jargon abbreviations and omits any explicit "Use when..." trigger guidance. It is recognizable but not comprehensive or trigger-rich, landing at the middle anchor across all dimensions.

Suggestions

Add an explicit "Use when..." clause naming concrete triggers (e.g., "Use when building declarative LM pipelines, auto-optimizing prompts with training data, or constructing RAG/agent systems with DSPy") to raise completeness and trigger_term_quality.

Expand jargon abbreviations into natural user terms — spell out "language model programs" and "retrieval-augmented generation" alongside DSPy so the description matches phrases users would actually say.

List a few more concrete actions (e.g., "compile signatures, bootstrap few-shot examples, evaluate pipelines") to move specificity from a compressed list toward multiple spelled-out capabilities.

Dimension	Reasoning	Score
Specificity	Quotes "declarative LM programs, auto-optimize prompts, RAG" name the domain and several actions, but they are compressed and jargon-heavy (LM, RAG) rather than comprehensively spelled out, matching anchor 2 rather than the multiple-concrete-actions bar of 3.	2 / 3
Completeness	It states what the skill does but includes no "Use when..." trigger clause; the rubric guideline caps completeness at 2 when explicit trigger guidance is missing.	2 / 3
Trigger Term Quality	Terms "DSPy", "auto-optimize prompts", and "RAG" are relevant, but "LM programs" is jargon and common natural variations are missing, fitting anchor 2 instead of the broad natural-term coverage of 3.	2 / 3
Distinctiveness Conflict Risk	Naming DSPy specifically gives it a distinct niche, but "auto-optimize prompts" and "RAG" overlap with many prompt-engineering/RAG skills, so it is only somewhat specific rather than clearly non-conflicting.	2 / 3
	Total	8 / 12 Passed

Validation

75%

Warnings & errors only

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

Validation — 12 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (595 lines); consider splitting into references/ and linking	Warning
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	12 / 16 Passed

Repository: NousResearch/hermes-agent
Path: optional-skills/mlops/research/dspy/SKILL.md
Commit: 4c9628e

Reviewed: about 13 hours ago

Table of Contents

Discovery Implementation Validation

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.