CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

maintaining-deferred-memory

Maintains deferred memory, which stores accumulated knowledge, analysis, and observations made by the agent over time. Use when saving knowledge (e.g. about a person, project, repository, system, or event). Do NOT use for managing skills, prompts, or for saving information that is already automatically stored in the message history.

Quality

85%

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

82%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.

A tightly written, highly actionable instruction-only skill: concrete file paths, config limits, formats, and a complete commit template with worked examples, organized in execution order and including explicit validation steps. Minor room remains to tighten light redundancy across sections and to consider splitting the commit-format detail into a reference file.

DimensionReasoningScore

Conciseness

The body is dense, prescriptive policy with no explanation of concepts Claude already knows, matching the level-4 'efficient; minor instances that could be trimmed' anchor — e.g. the don't-duplicate-recall point recurs across the intro, the NOT-list, and "Write analysis, not logs", and "Write the indexes" could be tightened. It is far from the padded verbosity of level 2 but not quite 'every token earns its place' at level 5.

4 / 5

Actionability

Fully concrete, copy-paste-ready guidance throughout: the config path "$MEMORY_DIR/.memfs.config.json" with named keys and defaults (maxDepth, maxFileCharacters), a full directory layout diagram, the exact two-field frontmatter block, link syntax like "[Acme](projects/acme.md)", and a complete git commit template with worked examples ("memory(self): Sarah wants one commit per learning"). Specific examples cover the common cases, matching the level-5 anchor; per the rubric's scoring note, absence of code in an instruction-only skill is not penalized when guidance is this actionable.

5 / 5

Workflow Clarity

Sections follow a logical execution order (check limits → layout → write indexes → design for disclosure → write analysis → commit) and include an explicit validation checkpoint with a feedback loop ("After adding or moving a file, trace the path a future request would take from root MEMORY.md to the file. If any step depends on knowing the file already exists, fix the index"), plus sequencing control for batch commits ("finish and commit one before editing for the next"). It stops short of level 5 because there is no consolidated step-by-step workflow or checklist framing, matching level 4's 'clear sequence with most checkpoints present'.

4 / 5

Progressive Disclosure

A single well-organized file with clear section headers and no nested or buried references; the skill's own design rules are consistently applied. At roughly 100 lines, the commit-format section is borderline content that could live in a one-level-deep reference, which is the minor organization gap of the level-4 anchor rather than the fully 'appropriately split' structure of level 5 (and the under-50-line exception does not apply).

4 / 5

Total

17

/

20

Passed

Description

82%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 that explicitly states what the skill does, when to use it with concrete example subjects, and when not to use it via explicit exclusions. Trigger terms are natural and the niche is well protected from conflicts. The only weakness is that the action list is limited to maintaining/storing rather than enumerating the skill's full range of operations.

DimensionReasoningScore

Specificity

"Maintains deferred memory, which stores accumulated knowledge, analysis, and observations made by the agent over time" names the domain and 1-2 concrete actions (maintains, stores) with example subjects ("person, project, repository, system, or event"), but does not list the several distinct operations of the level-4 anchor; it is clearly more concrete than the level-2 'names the domain only' anchor.

3 / 5

Completeness

It explicitly answers both what ("Maintains deferred memory, which stores accumulated knowledge, analysis, and observations") and when ("Use when saving knowledge (e.g. about a person, project, repository, system, or event)") with concrete trigger phrases, exactly matching the level-5 anchor and clearly above level 4's 'when could be more explicit'.

5 / 5

Trigger Term Quality

"Use when saving knowledge (e.g. about a person, project, repository, system, or event)" gives good natural keyword coverage a user would plausibly say, matching the level-4 anchor; common variations like "remember this" or "take notes" are missing, so it falls short of the comprehensive-synonym coverage of level 5.

4 / 5

Distinctiveness Conflict Risk

A clear niche (deferred memory) with explicit negative carve-outs — "Do NOT use for managing skills, prompts, or for saving information that is already automatically stored in the message history" — that actively disambiguate it from adjacent skills, giving minimal conflict risk as the level-5 anchor describes.

5 / 5

Total

17

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

relative_links

Relative link issues: 1 missing

Warning

Total

15

/

16

Passed

Repository

letta-ai/letta-code

Path

src/skills/builtin/maintaining-deferred-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.