CtrlK
BlogDocsLog inGet started
Tessl Logo

stream-builder

The framework-agnostic Stream builder (Steps 0-7 scaffold + enhance + Video audit), being extended to support app kinds beyond web. Use ONLY when the user explicitly names 'stream-builder' or '/stream-builder'. Web React / Next.js work - 'build me a chat app', 'scaffold', 'create a new ...', 'add Chat/Video/Feeds to this app', 'integrate Video', livestreaming, video conferencing, team messaging, direct messaging, social feed - defaults to the stream-react skill, not this one. Not auto-selected for generic web build requests.

87

2.42x
Quality

Does it follow best practices?

Impact

85%

2.42x

Average score across 3 eval scenarios

SecuritybySnyk

Advisory

Suggest reviewing before use

SKILL.md
Quality
Evals
Security

Quality

Content

77%

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 builder workflow with strong validation checkpoints, but it is somewhat verbose with repeated audit-skip and trust instructions, and two key referenced files (builder-ui.md, enhance.md) are missing from the bundle, breaking progressive disclosure for the code-generation and enhance flows.

Suggestions

Consolidate the Track F audit-skip guidance: it is stated in the intro three-flows list and repeated verbatim in the Start section — keep it in one place to save tokens.

Restore or repoint the broken references: Step 4 loads builder-ui.md and Track E points to enhance.md, but neither file is in the bundle — add them or redirect to existing references/* files so navigation does not dead-end.

Consider moving the detailed per-use-case Page Flow routing into a reference file (e.g. references/PAGE-FLOW.md) and linking out, keeping SKILL.md a lean overview.

DimensionReasoningScore

Conciseness

The body is mostly operational and assumes Claude's competence, but it repeats itself: the Track F audit-skip instruction appears both in the intro three-flows list and again verbatim in the Start section, and the verbatim Trust readout overlaps the full Install trust & integrity table. It could be tightened without losing clarity, matching the 'mostly efficient but could be tightened' anchor rather than the lean 3.

2 / 3

Actionability

Provides copy-paste-ready, executable commands throughout (npx shadcn@latest init ..., npm install <packages> --legacy-peer-deps, getstream env, the node icon-check script, npx tsc --noEmit, npx next build), matching the 'fully executable code/commands; copy-paste ready' anchor.

3 / 3

Workflow Clarity

Steps 0-7 are explicitly sequenced with a scaffold-order list, STOP-and-wait checkpoints (Step 2 theme pick, Task A.2 disclosure), and a validate-then-fix feedback loop in Step 5 (tsc --noEmit catches all errors, then next build, fix and re-run), plus a Reviewer checklist — matching the clear-sequence-with-explicit-validation anchor.

3 / 3

Progressive Disclosure

Structure is mostly good (overview SKILL.md pointing one level deep to references/<Product>.md, *-blueprints.md, CROSS-PRODUCT.md, and use-cases/ recipes, all present in the bundle), but two core load instructions point to files absent from the bundle: Step 4 loads builder-ui.md and Track E references enhance.md — neither exists at the repo root. Broken navigation for core flows keeps this below the well-signaled one-level-deep 3 anchor.

2 / 3

Total

10

/

12

Passed

Description

100%

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 that concretely states capabilities, gives explicit trigger guidance, and proactively disambiguates from the sibling stream-react skill. It is slightly verbose with meta-commentary about ongoing extension, but every dimension is explicitly addressed.

DimensionReasoningScore

Specificity

Names multiple concrete actions ('Steps 0-7 scaffold + enhance + Video audit') plus a specific product set (Chat/Video/Feeds) and app kinds (livestreaming, video conferencing, team messaging, direct messaging, social feed), matching the 'lists multiple specific concrete actions' anchor.

3 / 3

Completeness

Answers both what ('framework-agnostic Stream builder ... scaffold + enhance + Video audit') and when with an explicit 'Use ONLY when the user explicitly names stream-builder or /stream-builder' clause, satisfying the what-and-when anchor.

3 / 3

Trigger Term Quality

Enumerates natural user phrasings ('build me a chat app', 'scaffold', 'add Chat/Video/Feeds', 'integrate Video', livestreaming, etc.) and explicitly distinguishes which trigger this skill vs. the sibling, giving strong coverage of terms users actually say.

3 / 3

Distinctiveness Conflict Risk

Carves a clear niche ('framework-agnostic', 'app kinds beyond web') and explicitly disambiguates from stream-react ('defaults to the stream-react skill, not this one. Not auto-selected for generic web build requests'), making wrong-skill triggering unlikely.

3 / 3

Total

12

/

12

Passed

Validation

75%

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

Validation12 / 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: 2 missing, 1 deeper-than-1-level, 3 suspicious

Warning

referenced_paths_exist

Referenced path issues: 2 deeper-than-1-level

Warning

Total

12

/

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.