chroma

Open-source embedding database for AI applications. Store embeddings and metadata, perform vector and full-text search, filter by metadata. Simple 4-function API. Scales from notebooks to production clusters. Use for semantic search, RAG applications, or document retrieval. Best for local development and open-source projects.

Quality

70%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./backend/cli/skills/llm-tools/chroma/SKILL.md

Quality

Content

57%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 content is highly actionable with executable code throughout, but it is verbose, duplicates the bundled integration reference inline, and lacks validation checkpoints for its batch and destructive operations. Navigation to the existing reference file is not signaled, capping progressive disclosure.

Suggestions

Replace the inline LangChain and LlamaIndex sections with a one-line pointer to references/integration.md (e.g. '## Integrations — See [integration.md](references/integration.md)') to cut tokens and surface the reference.

Add validation/verification steps to the batch add and delete workflows (e.g. confirm counts after add, verify remaining documents after delete) to lift workflow clarity above the destructive/batch cap of 3.

Trim marketing padding (GitHub stars/forks metrics, repeated embedding-function setup boilerplate) and consolidate the duplicated embedding-function examples into a single reference.

Dimension	Reasoning	Score
Conciseness	The body is noticeably verbose for a skill overview — it inlines full LangChain and LlamaIndex integration sections that duplicate the bundled references/integration.md, restates embedding-function setups multiple times, and pads with marketing metrics (GitHub stars, forks), several unnecessary explanations Claude does not need.	2 / 5
Actionability	Almost every operation is shown as complete, copy-paste-ready executable Python covering create/add/query/get/update/delete, persistence, multiple embedding functions, metadata filters, and server mode, matching the fully-executable anchor.	5 / 5
Workflow Clarity	Sections are organized as a labeled operation catalog rather than a sequenced workflow, and despite covering batch add and destructive delete operations, there are no validation or verification checkpoints, so the destructive/batch cap of 3 applies.	3 / 5
Progressive Disclosure	There is some section structure and one real reference file (references/integration.md), but its existence is never signaled from the body — instead the LangChain/LlamaIndex content is duplicated inline — so references are buried and content that belongs in a separate file is inlined.	3 / 5
	Total	13 / 20 Passed

Description

83%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 well-constructed: it states concrete capabilities, gives explicit 'Use for...' trigger guidance, and positions Chroma against a clear niche. Minor gaps in action coverage and synonyms keep specificity and trigger quality just below the top anchor.

Dimension	Reasoning	Score
Specificity	Lists several specific concrete actions — 'Store embeddings and metadata', 'perform vector and full-text search', 'filter by metadata' — with only minor coverage gaps (no mention of update/delete operations), matching the 'several specific actions' anchor.	4 / 5
Completeness	Explicitly answers what it does ('Store embeddings and metadata, perform vector and full-text search, filter by metadata') and when to use it ('Use for semantic search, RAG applications, or document retrieval'), satisfying both halves with concrete trigger phrases.	5 / 5
Trigger Term Quality	Includes strong natural user phrases like 'semantic search', 'RAG applications', 'document retrieval', and 'metadata filtering'; a few common synonyms or explicit data-type terms are missing, so it sits at good-but-not-comprehensive coverage.	4 / 5
Distinctiveness Conflict Risk	The 'Best for local development and open-source projects' and 'embedding database' framing carves a clear niche distinct from managed vector stores, with only minor overlap risk against other RAG/vector-database skills.	4 / 5
	Total	17 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/llm-tools/chroma/SKILL.md
Commit: fcd2e11

Reviewed: 2 days 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.