CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

add-agent-memory

Wire Redis Agent Memory into an existing agent with redis-iris so it remembers users across sessions. Asks five questions (what to remember, what to share, what never to keep, how long to keep it, whether users can be forgotten), writes the answers to service.yaml, provisions the service from the terminal with `iris agent-memory service create`, then wires recall and write-back into the agent's turn in any framework. Use when the agent forgets users, uses InMemoryStore or a checkpointer only, keeps memory in a dict or local DB, or the user asks for long-term memory, recall, or right-to-erasure.

Quality

94%

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

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.

The body is an unusually actionable, well-sequenced runbook: copy-paste commands, executable code, explicit validation gates, and a complete error-recovery table. Its one structural weakness is monolithic presentation — specialized material (identity tokens, console provisioning, migration) that belongs in reference files is inlined in SKILL.md.

Suggestions

Move the Identity-tokens precedence discussion, the Redis Cloud console walkthrough, and the full error-code table into a references/ file (e.g. references/provisioning.md), keeping a one-line pointer plus the three or four most common error codes in SKILL.md.

Extract the "Already on Redis Agent Memory" migration section into its own reference file and link it from step 1, since it applies only to a subset of projects.

Trim the overlap between the six-item "Run the CLI as an agent" loop and the error-code table — keep the loop about process (dry-run first, stop on first failure) and let the table alone carry per-code actions.

DimensionReasoningScore

Conciseness

The body is dense and product-specific throughout — no explanations of concepts Claude already knows, and the five-question table and error-code table are token-efficient. Minor instances could be trimmed (the Identity-tokens paragraph and the numbered CLI loop partially overlap the error-code table), so it sits at efficient-with-minor-trims rather than every-token-earns-its-place. Not 3: there is no genuine padding or generic explanation.

4 / 5

Actionability

Fully executable guidance: exact CLI commands ("iris --json --no-input agent-memory service create --from service.yaml --dry-run"), a complete commented service.yaml, runnable Python ("AgentMemory.from_env(assistant_id=\"<agent-name>\")", "memory.recall(question, owner=user_id, shared=[floor])", "convo.user(question)"), and a named integration matrix ("attach(memory, graph) from `redis_iris.integrations.langgraph`", extra "redis-iris[langgraph]"). Not 4: the common cases — ask, configure, provision, wire, verify — are each covered with copy-paste-ready material.

5 / 5

Workflow Clarity

Six clearly sequenced steps with explicit validation checkpoints and feedback loops: MCP fact-check before code ("Never guess framework APIs from memory"), dry-run gating ("If `missing` is not empty, stop and hand off"), post-provision "service check", an 8-item verification checklist including failure paths ("Unset a variable and confirm `ConfigurationError` names it"), error recovery ("re-run with `--reuse --write-env`"), and destructive-op guardrails ("Never delete to retry"). Not 4: validation is explicit at every phase, not mostly present.

5 / 5

Progressive Disclosure

No bundle files exist and the body references none; the Identity-tokens deep-dive, the Redis Cloud console walkthrough, the full error-code table, and the "Already on Redis Agent Memory" migration guide are all inlined in a ~200-line monolith — content that clearly belongs in separate files is inline. Not 2: the internal section structure (six numbered steps, tables, clear subsections) is strong rather than minimal; not 4: at this size, zero one-level-deep references is more than a minor organization gap.

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.

The description is exemplary: it enumerates the full workflow in concrete, imperative voice and pairs it with an explicit Use-when clause covering symptoms, current implementations, and user requests. It is dense rather than padded, and every clause carries actionable information.

DimensionReasoningScore

Specificity

Lists multiple concrete actions covering the entire workflow — "Asks five questions (what to remember, what to share, what never to keep, how long to keep it, whether users can be forgotten)", "writes the answers to service.yaml", "provisions the service from the terminal with `iris agent-memory service create`", "wires recall and write-back into the agent's turn in any framework" — matching the comprehensive-coverage anchor. Not 4: there are no coverage gaps across ask, configure, provision, and wire.

5 / 5

Completeness

Explicitly answers both "what" (wire Redis Agent Memory, ask five questions, write service.yaml, provision via CLI, wire recall and write-back) and "when" via a concrete "Use when..." clause listing five trigger conditions. Not 4: the when-clause is fully explicit with concrete trigger phrases rather than merely present.

5 / 5

Trigger Term Quality

Covers natural phrases users would actually say across symptom triggers ("the agent forgets users"), implementation-state triggers ("uses InMemoryStore or a checkpointer only, keeps memory in a dict or local DB"), and request triggers ("long-term memory, recall, or right-to-erasure"), with synonyms (forgets/remembers/recall). Not 4: coverage spans all trigger categories comprehensively; file extensions are not applicable to this domain.

5 / 5

Distinctiveness Conflict Risk

Clear niche — Redis Agent Memory via redis-iris — with distinct triggers like "InMemoryStore or a checkpointer only" and "right-to-erasure" that no closely related skill would claim. Not 4: overlap risk is minimal, not minor.

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-agent-memory/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.