CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

add-context-gateway

Put the MCP servers an existing agent uses behind one Redis Iris Context Gateway with redis-iris, so the agent connects to one MCP endpoint with one agent key instead of configuring each server and its secret. Asks whether the Gateway is in Redis Cloud or self-managed, then five questions (which MCP servers and secrets the agent has today, how many tools it sees, which read-only tools may be cached, which tools return large results, whether the framework speaks MCP itself). Self-managed, it provisions the Gateway and its Upstreams from the terminal with `iris context-gateway` and sets cache and compression policies. In Redis Cloud the developer creates them in the console. Then it points the agent at the Gateway in any framework and verifies it with `tools list` and `tools call`. Use when the agent configures several MCP servers one by one, carries one secret per server, loads too many tool schemas into the prompt, or the user asks for Context Gateway, one MCP endpoint, tool caching, or result handles.

Quality

94%

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

85%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.

An exceptionally actionable and well-sequenced operational skill: copy-paste commands, complete examples, explicit verification, and error-recovery guidance covering both deployment modes. Its main weaknesses are the absence of any progressive disclosure — all reference material is inlined in one long file — and inline version-sensitive workarounds that will age.

Suggestions

Move the data-model specification (level/member tables, index-selection table, example JSON) and the handles reference (the seven handle tools and their call signatures) into files under references/, leaving short linked sections in SKILL.md.

Collect version-sensitive notes (the tool_not_found name-splitting workaround for older redis-iris releases, 'Redis Cloud can't embed queries yet', known issue MOD-20735) into a clearly labeled 'Compatibility and known issues' section so the main flow does not age with releases.

Trim the 'Gateway switches and database' sizing guidance to the commands and the error signals to check (skipped_low_memory / out_of_memory), cutting the derivations of share arithmetic.

DimensionReasoningScore

Conciseness

The body is dense and operational with no explanations of concepts Claude already knows, but version-sensitive notes ('the installed release splits tool names at the first `.`: upgrade redis-iris', 'Redis Cloud can't embed queries yet') sit inline rather than in a compatibility/deprecated section, and a few passages (identity token precedence, memory-share sizing) could be tightened. This matches 'efficient with minor instances that could be trimmed' rather than 'every token earns its place'.

4 / 5

Actionability

Fully executable throughout: exact CLI invocations with flags (`iris --json --no-input context-gateway service create --name support --tool-exposure-mode full --write-env .env`), a complete copy-paste data-model JSON, an error-code table mapping each `error.code` to who acts and what to do, and verification commands with expected outputs. Not score 4: it covers both Redis Cloud and self-managed paths and both exposure modes with concrete, runnable commands for the common cases.

5 / 5

Workflow Clarity

Six numbered sections (Discover → Get current facts → Provision → Repoint → Verify → Report) with explicit validation checkpoints: 'Read before you write' preflight via `service list`, a 9-item verification checklist with expected results (cache-hit `_meta`, `handle_not_found` on a released handle), and feedback loops for error recovery (412 `revision_conflict` → re-read before retrying, 503 → retry, `resource_exists` → `--reuse details.gatewayId`). Matches the anchor for clear sequence with explicit validation and recovery loops.

5 / 5

Progressive Disclosure

There are no bundle files at all (no references/, scripts/, or assets/), and roughly 150 lines of reference material — the data-model level/member and index tables, database requirements, and the handle-tools table — are inlined in a single ~300-line SKILL.md where separate reference files would keep the body an overview. This mirrors the anchor '[200 lines of API reference that could be in a separate file]'. Not score 4: content that clearly belongs in separate files is inline; not score 2: the file is well-sectioned, with working internal cross-references ('see "Write the data model"') making it navigable.

3 / 5

Total

17

/

20

Passed

Description

100%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.

An exemplary description: third-person, concrete about every capability, explicit about both deployment modes, and ending in a rich 'Use when...' clause with natural trigger phrases and synonyms. It is long, but every clause carries distinguishing information rather than padding.

DimensionReasoningScore

Specificity

Multiple concrete actions with no padding: 'provisions the Gateway and its Upstreams from the terminal with `iris context-gateway`', 'sets cache and compression policies', 'points the agent at the Gateway in any framework', 'verifies it with `tools list` and `tools call`'. Coverage spans discovery (the five questions), provisioning, policy, repointing, and verification — comprehensive rather than 'minor gaps', so anchor 5 over anchor 4.

5 / 5

Completeness

Both halves are explicit: the opening sentences state what the skill does (put MCP servers behind one Gateway, provision, set policies, repoint, verify), and a dedicated 'Use when...' clause lists concrete trigger conditions. This matches the anchor-5 example's structure of clear 'what' plus explicit 'when' with concrete triggers, exceeding anchor 4's 'when could be more explicit'.

5 / 5

Trigger Term Quality

Natural user phrasings plus synonyms across the domain: 'configures several MCP servers one by one', 'carries one secret per server', 'loads too many tool schemas into the prompt', 'Context Gateway, one MCP endpoint, tool caching, or result handles'. Within this product's domain no meaningful natural term is missing, matching comprehensive coverage with synonyms rather than 'a few natural terms missing'.

5 / 5

Distinctiveness Conflict Risk

A clear niche (Redis Iris Context Gateway, redis-iris, MCP endpoint consolidation) with product-specific triggers ('Context Gateway', 'one MCP endpoint', 'result handles') that generic MCP-configuration skills would not match. Minimal conflict risk, matching anchor 5 rather than anchor 4's 'minor overlap risk with closely related skills'.

5 / 5

Total

20

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

redis/agent-skills

Path

plugins/redis-development/skills/add-context-gateway/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.