initializing-memory

Comprehensive guide for initializing or reorganizing agent memory. Load this skill when running /init, when the user asks you to set up your memory, or when you need guidance on creating effective memory files.

Quality

72%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./src/skills/builtin/initializing-memory/SKILL.md

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.

This is a highly actionable, tightly sequenced workflow skill: every step has executable commands, real bundled scripts, validation checkpoints, and explicit recovery paths for failure cases. Its main structural weakness is that all of the detail lives in SKILL.md itself — the workflow-authoring guidance, schema code, and verification checklist would sit better in one-level-deep reference files.

Suggestions

Move the Step 7 workflow-authoring rules and the schema JS into a references/workflow-analysis.md, keeping a short summary and the schema snippets' shape inline in SKILL.md.

Extract the Step 9 verification checklist into references/verification.md (or a script alongside history-coverage.mjs) so the main file reads as the overview-plus-commands layer.

Dimension	Reasoning	Score
Conciseness	The body is dense and skill-specific throughout — validation rules, cohort budgets, provenance gates — with no explanations of concepts Claude already knows. A few passages could be trimmed (e.g., the philosophical asides in "Identity and continuity" and the multi-clause sentences that restate the same constraint), which keeps it below the "every token earns its place" anchor.	4 / 5
Actionability	Nearly every step is copy-paste executable: "letta trajectories detect", "git shortlog -sn --all \| head -5", the jq manifest summary, "node <SKILL_DIR>/scripts/prepare-history.mjs --export ... --out ...", the complete runnable workflow-schema JS, the violation-finding find pipeline, and the full git commit sequence with flags. Both referenced scripts (scripts/prepare-history.mjs, scripts/history-coverage.mjs) are real bundle files invoked with explicit arguments.	5 / 5
Workflow Clarity	A ten-step sequence (inspect → detect → identify → ask → export → research → analyze → curate → verify → commit) with explicit validation checkpoints and feedback loops: the coverage script writes unread sessions to retry.json for a follow-up workflow and is re-run with both journals, Step 9 is a full verification checklist (structure, duplication, granularity, persona), and Step 10 verifies the commit with a second "git status" before reporting. Error paths (workflow tool unavailable, incomplete coverage, blocked commit) are each handled explicitly.	5 / 5
Progressive Disclosure	The body is well-sectioned with headers and correctly points at the two real bundle scripts, but the SKILL.md is a ~210-line monolith that carries all detail inline — the Step 7 workflow-authoring rules and full schema code, the export/cohort mechanics, and the Step 9 checklist are material that belongs in reference files for a skill this size. There is no references/ directory, so the overview-points-to-detail pattern is only partially realized.	3 / 5
	Total	17 / 20 Passed

Description

60%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 has strong, explicit trigger guidance and occupies a distinct niche, but the capability statement is generic ("Comprehensive guide for...") and written partly in second person, so it never tells the reader what the skill concretely does. Rewriting the first sentence as concrete third-person actions would lift both specificity and completeness.

Suggestions

Replace "Comprehensive guide for initializing or reorganizing agent memory" with the concrete actions the skill performs, e.g. "Detects and analyzes coding-agent session history, structures an indexed agent memory directory, writes persona and preference files, and commits the result to $MEMORY_DIR."

Use third person and drop the second-person clause: e.g. "Use when running /init or when the user asks to set up, initialize, or reorganize agent memory."

Add natural trigger synonyms such as "organize my memory", "memory setup", and "MEMORY.md" to broaden keyword coverage.

Dimension	Reasoning	Score
Specificity	"Comprehensive guide for initializing or reorganizing agent memory" names the domain but the actions are generic restatements of it — it never states what the skill actually does (detect session history, research the codebase, write indexed memory files, commit). Base level is a 2 ("names the domain but actions are minimal or generic"), reduced by 1 per the rubric's voice rule for the second-person phrasing "when the user asks you to set up your memory, or when you need guidance".	1 / 5
Completeness	Both parts are present: a "what" ("guide for initializing or reorganizing agent memory") and an explicit multi-trigger "when" clause ("Load this skill when running /init, when the user asks you to set up your memory, or when you need guidance..."). The "when" is concrete, but the "what" stays vague — it does not name the concrete capabilities behind the "guide" — so it does not reach the fully-explicit anchor 5.	4 / 5
Trigger Term Quality	Includes natural trigger phrases users would actually say — "running /init", "set up your memory", "creating effective memory files" — plus "initializing" and "reorganizing". A few natural synonyms are missing (e.g., "organize my memory", "MEMORY.md", "memory setup"), so it falls just short of the comprehensive-synonyms anchor.	4 / 5
Distinctiveness Conflict Risk	"Agent memory" initialization/reorganization is a clear niche with distinct triggers ("set up your memory", "MEMORY" files), giving minimal conflict with unrelated skills. Minor overlap risk remains with closely related skills — "/init" also naturally triggers repository/CLAUDE.md initialization skills — keeping it below the minimal-conflict anchor.	4 / 5
	Total	13 / 20 Passed

Validation

93%

Warnings & errors only

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

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 1 missing	Warning

	Total	15 / 16 Passed

Repository: letta-ai/letta-code
Path: src/skills/builtin/initializing-memory/SKILL.md
Commit: d4d1db7

Reviewed: 5 days ago

Table of Contents

Discovery Implementation Validation

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.