CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

agents-langchain

Framework for building LLM-powered applications with agents, chains, and RAG. Supports multiple providers (OpenAI, Anthropic, Google), 500+ integrations, ReAct agents, tool calling, memory management, and vector store retrieval. Use for building chatbots, question-answering systems, autonomous agents, or RAG applications. Best for rapid prototyping and production deployments.

Quality

73%

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 ./cli-tool/components/skills/ai-research/agents-langchain/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

56%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.

The body is a broad, concrete, well-sectioned tour of LangChain with a clear quick start, but it is undermined by mixed-era and partly non-executable code, marketing padding, and heavy duplication of material that already lives in the reference files. Tightening the body to an overview plus quick start and pushing catalog detail into the existing references would fix most issues.

Suggestions

Cut the GitHub-stars/contributors metrics block and the 'Performance benchmarks' table — they are unverifiable marketing filler that spends context without changing what Claude does.

Fix the non-executable snippets: WebBaseLoader.load() takes no URL argument, create_tool_calling_agent needs a ChatPromptTemplate not a string, and the LLMChain/ConversationBufferMemory/RetrievalQA examples are deprecated — pick one LangChain version and make every example runnable against it.

Replace the inline Vector stores, Document loaders, and Text splitters sections with pointers to references/rag.md and references/integration.md (as already done for Agents/RAG/Integration), keeping only one compact example of each in the body.

DimensionReasoningScore

Conciseness

Most of the body is dense, useful code, but there is real padding: a marketing metrics block ("119,000+ GitHub stars", "272,000+ repositories", "3,800+ contributors"), a speculative "Performance benchmarks" table ("~1-2s", "~10-30s" with no methodology), and ~150 lines of vector-store/loader/splitter catalog content that duplicates references/rag.md and references/integration.md. This sits between anchor 2 ('several unnecessary padded sections') and anchor 3 ('mostly efficient with some unnecessary explanation') — the code core is efficient, so 3 fits better than 2, but it is not anchor 4's 'minor instances'.

3 / 5

Actionability

Code examples dominate and installation/basic-usage/agent snippets are runnable, but several snippets are not executable as written: `loader.load("https://...")` (WebBaseLoader.load() takes no arguments), `create_tool_calling_agent(llm=..., tools=..., prompt="Answer questions...")` (requires a ChatPromptTemplate, not a string), and the deprecated `LLMChain`/`ConversationBufferMemory`/`RetrievalQA`/`ConversationalRetrievalChain` APIs mixed with 1.x-era `create_agent` while claiming "Version: 0.3+". These are more than anchor 4's 'minor gaps', pulling it down to anchor 3, though it is far above anchor 2's high-level hints.

3 / 5

Workflow Clarity

The quick start is a clear sequence (install → basic LLM usage → agent) and the RAG pipeline is explicitly numbered 1-6 with a runnable end-to-end path; decision routing is aided by 'When to use LangChain' and the LangChain vs LangGraph table. No validation checkpoints appear, but no destructive or batch operations are involved, so this matches anchor 4 ('clear sequence with most checkpoints present; minor validation gaps') rather than anchor 5's explicit validate/fix/retry loops or anchor 3's mere step list.

4 / 5

Progressive Disclosure

The bundle structure is good — references/agents.md, rag.md, and integration.md all exist, are one level deep with no nested references, and are clearly signaled in a '## References' section with one-line descriptions. However, the body inlines large API-catalog sections (Vector stores, Document loaders, Text splitters, and the full basic RAG pipeline) that duplicate the reference files, matching anchor 3's 'content that should be separate is inline' rather than anchor 4's 'most content appropriately placed'.

3 / 5

Total

13

/

20

Passed

Description

80%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.

A strong description: concrete capability list, explicit 'Use for...' trigger guidance, and good natural keywords. Its main weakness is that the brand name is absent from the description text, creating overlap risk with other LLM-framework skills, plus minor coverage gaps (document loaders, text splitters).

Suggestions

Include the framework name in the description itself (e.g., 'LangChain framework for building...') so it is distinguishable from LlamaIndex/Haystack/LangGraph skills when only the description is matched.

DimensionReasoningScore

Specificity

The description lists many concrete capabilities ("agents, chains, and RAG", "ReAct agents, tool calling, memory management, and vector store retrieval", "multiple providers (OpenAI, Anthropic, Google)"), but coverage has minor gaps — document loaders, text splitters, and streaming, all documented in the body, are omitted. This matches anchor 4 ('several specific actions; minor gaps') better than anchor 5's comprehensive coverage, and is well above anchor 3's 1-2 actions.

4 / 5

Completeness

Both halves are explicit: the 'what' is concrete ("Framework for building LLM-powered applications with agents, chains, and RAG... tool calling, memory management, and vector store retrieval") and the 'when' is an explicit trigger clause ("Use for building chatbots, question-answering systems, autonomous agents, or RAG applications") with concrete use cases. This matches anchor 5's example pattern; it is not anchor 4, where the 'when' is only weakly specified.

5 / 5

Trigger Term Quality

Good natural keyword coverage: "chatbots", "question-answering systems", "autonomous agents", "RAG", "tool calling", "vector store retrieval" are phrases users would plausibly say. A few natural variants are missing ("LangChain" itself, "LLM app/framework", "retrieval-augmented generation" spelled out, "conversational AI"), so it sits at anchor 4 rather than anchor 5's comprehensive synonym coverage, but clearly above anchor 3's partial coverage.

4 / 5

Distinctiveness Conflict Risk

The description never names the framework it describes — "LangChain" appears only in the name field, not in the description text — so "Framework for building LLM-powered applications with agents, chains, and RAG" could equally trigger skills for LlamaIndex, Haystack, or LangGraph. That is anchor 3 ('somewhat specific but could still overlap with similar skills'); it lacks anchor 4's minimal-conflict distinctiveness, though it is more specific than anchor 2's broad domain claim.

3 / 5

Total

16

/

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

metadata_field

'metadata' should map string keys to string values

Warning

Total

15

/

16

Passed

Repository

davila7/claude-code-templates

Path

cli-tool/components/skills/ai-research/agents-langchain/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.