CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

add-lens

Add Redis Iris Lens tracing to an existing AI agent through redis-iris, so every agent turn is one trace in the Lens UI with its LLM calls, tokens, tool calls, memory and document retrievals, and which of them entered the prompt. Detects how the agent runs (LangGraph or ADK attach, workflows.turn, another framework, or a plain loop), installs redis-iris[llm-tracing], calls lens.init(agent_id=...), opens one lens.turn() per turn where Iris does not open it already, and covers coexistence with an OpenTelemetry setup or LLM instrumentation the app already has. Use when the user asks to add Lens, tracing, observability, or per-turn traces to their agent, or asks why their Lens traces are missing.

Quality

90%

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

SKILL.md
Quality
Evals
Security

Quality

Content

88%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 an exceptionally actionable integration runbook: every step carries executable code or a decision table, and the verify step provides a genuine diagnostic feedback loop. Its only deductions are mild repetition of the agent_id/session_id rationale and the absence of any bundle split for a fairly long single-file skill.

DimensionReasoningScore

Conciseness

Dense with product-specific facts Claude cannot know ("lens.turn() is a plain context manager and works around await"; "Its return value only says an endpoint is set... check lens.instrumentation() instead") with no explanation of basic concepts. Minor redundancy — the agent_id/session_id visibility rationale is repeated in steps 1, 4, and 7 — so it sits at the 4 anchor (efficient, minor trimmable instances) rather than the every-token-earns-its-place 5 anchor.

4 / 5

Actionability

Fully executable throughout: per-package-manager install commands ("uv add 'redis-iris[llm-tracing]'", "poetry add"), a .env template, copy-paste lens.init() with instrumentation check, the lens.turn() wrapper, lens.span/retrieval/prompt_assembly examples, lens.inject()/continue_from for process hops, lens.shutdown(timeout_millis=5_000), plus a shape-matching table naming the exact run call per framework (app.invoke, Runner.run, runner.run_async, agent.run, kickoff). Matches the 5 anchor: copy-paste ready and covers the common cases.

5 / 5

Workflow Clarity

Eight clearly sequenced steps (Discover → MCP facts → Install → Wire → Close gaps → Coexist → Verify → Report) with an explicit validation phase: expected values ('lens': 'on'), a Value→Cause→Fix troubleshooting table forming a feedback loop ("A value is wrong? Read the cause from lens.instrumentation()"), and a stats/UI checklist (iris.lens.spans_exported should grow, export_failures stay 0). Clear match to the 5 anchor; no destructive/batch operations so no cap applies.

5 / 5

Progressive Disclosure

No bundle files exist (references/, scripts/, assets/ absent), so all ~178 lines are inline — well-sectioned numbered steps and two operational tables, with the single external pointer clearly signaled ("Full reference: Lens (docs/core/lens.md in redis-iris)"). Good structure with minor organization gaps matches the 4 anchor; not 5 because no content is split across files and the <50-line simple-skill exception does not apply; not 3 since nothing that clearly belongs in a separate file is inlined.

4 / 5

Total

18

/

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 description: it states the full capability set in concrete, third-person terms and includes an explicit, multi-trigger "Use when" clause covering both the happy path and a debugging entry point. The only weaknesses are a couple of missing natural synonyms (monitoring, instrumentation) and mild trigger-word overlap with generic observability skills.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — "Detects how the agent runs (LangGraph or ADK attach, workflows.turn, another framework, or a plain loop), installs redis-iris[llm-tracing], calls lens.init(agent_id=...), opens one lens.turn() per turn... covers coexistence with an OpenTelemetry setup" — plus what the trace contains ("LLM calls, tokens, tool calls, memory and document retrievals"). Comprehensive coverage matching the 5 anchor; also consistently third-person, so no voice penalty.

5 / 5

Completeness

Explicitly answers both: what ("Add Redis Iris Lens tracing to an existing AI agent through redis-iris, so every agent turn is one trace in the Lens UI with its LLM calls, tokens, tool calls...") and when, with concrete trigger phrases in the "Use when..." clause. Clear match to the 5 anchor; not 4 because the when-clause is specific and includes a failure-mode trigger.

5 / 5

Trigger Term Quality

Explicit natural triggers: "Use when the user asks to add Lens, tracing, observability, or per-turn traces to their agent, or asks why their Lens traces are missing" — good synonym coverage (Lens, tracing, traces, observability). A few natural phrasings users would say are missing (e.g., "monitoring", "instrument my agent"), keeping it just below the comprehensive 5 anchor.

4 / 5

Distinctiveness Conflict Risk

Clear niche (Redis Iris Lens, per-turn agent tracing) with distinctive tokens ("Lens", "redis-iris[llm-tracing]", "lens.turn()"), but the bare trigger words "tracing" and "observability" could also fire for a generic OpenTelemetry/observability skill — minor overlap risk with closely related skills, matching the 4 anchor rather than the minimal-conflict 5 anchor.

4 / 5

Total

18

/

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-lens/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.