CtrlK
BlogDocsLog inGet started
Tessl Logo

stream-docs

Search live Stream SDK documentation for Chat, Video, Feeds, and Moderation. Look up how a Stream React/iOS/Android/Node/Flutter/Unity/Angular hook, component, or method works. Answer how-to questions about any Stream SDK across every framework and version. Triggers on 'docs', 'documentation', explicit SDK tokens (Chat React, Video iOS, Feeds Node, Moderation), and 'how do I ... in <framework>' phrasing. Answers come verbatim from getstream.io with citations - no CLI required.

67

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

77%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 highly actionable, well-sequenced docs-search workflow with strong validation and recovery. The main weaknesses are verbosity in the honesty-rules section and the absence of bundle files to offload the large routing/mapping tables.

Suggestions

Trim the 'Never say' bad-examples table and meta-narration in the Honesty rules to the essential rules; Claude can infer most of the leak-internals cases from a short principle.

Move the static reference tables (slug-name patterns, npm-package→product mapping, framework-name normalization) into a references/ file (e.g. SDK_MAP.md) linked from Step 1, reducing SKILL.md size and improving progressive disclosure.

The skill ships no bundle files but the body references ../stream/RULES.md and ../stream/SKILL.md; add a one-line note clarifying these are sibling pack skills, not bundle files of this skill, so navigation expectations are explicit.

DimensionReasoningScore

Conciseness

Mostly efficient for a complex routing skill, but the large 'Never say' bad-examples table and verbose honesty rationale explain behavior Claude could largely infer and could be tightened.

3 / 5

Actionability

Fully executable guidance: exact grep probes, exact URLs, verbatim fetch prompts, precise citation format, and concrete version 200/404 handling covering the common cases.

5 / 5

Workflow Clarity

Steps 1–4 are clearly sequenced with sub-steps and explicit validation checkpoints (verify URLs after fetch, re-fetch once with a stricter prompt, stop if still empty) plus 404 recovery feedback loops.

5 / 5

Progressive Disclosure

No bundle files ship with the skill, and large reference tables (slug patterns, package→product mappings, framework normalization) are inlined rather than split into navigable detail files, though section headers provide reasonable structure.

3 / 5

Total

16

/

20

Passed

Description

88%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, specific description that clearly states capabilities and explicit trigger conditions. Minor room to broaden natural-language trigger synonyms beyond SDK tokens.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — search live docs, look up hooks/components/methods, answer how-to questions, return verbatim answers with citations — giving comprehensive coverage rather than just a domain name.

5 / 5

Completeness

Explicitly answers both 'what' (search/lookup/answer from Stream SDK docs) and 'when' with concrete trigger phrases ('Triggers on 'docs', 'documentation', explicit SDK tokens...').

5 / 5

Trigger Term Quality

Good natural-term coverage ('docs', 'documentation', 'how do I ... in <framework>') plus explicit SDK tokens, though it leans on technical tokens more than a full synonym set users would naturally say.

4 / 5

Distinctiveness Conflict Risk

Clear niche tied to named Stream products/frameworks with explicit SDK tokens that reduce wrong-skill triggering, though the generic 'search documentation' framing carries minor overlap risk.

4 / 5

Total

18

/

20

Passed

Validation

81%

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

Validation13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

metadata_version

'metadata.version' is missing

Warning

relative_links

Relative link issues: 3 suspicious

Warning

Total

13

/

16

Passed

Repository
GetStream/agent-skills
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.