memory

Expertise in maintaining persistent bot memory, synchronizing with previous sessions via the Task Ledger, and preserving decision logs.

Quality

59%

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 ./tools/gemini-cli-bot/.gemini/skills/memory/SKILL.md

Quality

Content

68%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 efficient and highly actionable, with concrete commands, a copy-paste memory template, and explicit pruning limits. The main weaknesses are workflow-level: the Phase 0/Phase 6 numbering references an unstated larger process, key modes are undefined, and the destructive pruning operation lacks a validation step.

Suggestions

Define 'Scheduled Mode' vs 'Interactive Mode' and what 'the trigger' is, or reference where those are specified, since the workflow depends on them.

Add a validation step before pruning (e.g., confirm the ledger parses and the retained entries are the most recent) since discarding all but the last 50 tasks / 20 decisions is a destructive batch operation.

Resolve the phase numbering gap (Phase 0 and Phase 6 with no Phase 1-5 in this skill) by either renumbering locally or pointing to the workflow document that defines the full phase sequence.

Dimension	Reasoning	Score
Conciseness	The body is lean: it explains nothing Claude already knows, and the embedded lessons-learned.md template plus concrete pruning limits (50 tasks, 20 entries) are all substance. It falls one notch below anchor 5 due to minor redundant framing such as 'Once your investigation and implementation are complete'.	4 / 5
Actionability	Concrete guidance throughout: an exact file path (tools/gemini-cli-bot/lessons-learned.md), real commands (gh pr view, gh issue view), and a copy-paste-ready markdown template with example rows. It does not reach anchor 5 because 'Scheduled Mode', 'Interactive Mode', and 'the trigger' are referenced without being defined.	4 / 5
Workflow Clarity	Phase 0 (mandatory start) and Phase 6 (mandatory end) give a sequence with a 'Verify State' checkpoint, but the phase numbering implies phases 1-5 that are never defined, and the pruning step (a destructive/batch operation discarding all but the last 50 tasks and 20 decisions) has no validation. Per the destructive/batch guideline, workflow clarity is capped at 3.	3 / 5
Progressive Disclosure	A single-file skill with no nested references and a clean section hierarchy (Goal, Memory Structure, Rituals, Delegation), so navigation is easy. It scores 4 rather than 5 because the body is ~85 lines (above the under-50 simple-skill exemption) with the full memory template inlined where a separate reference file could be considered.	4 / 5
	Total	15 / 20 Passed

Description

50%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 states a clear 'what' with three named capability areas, but it has no 'Use when...' trigger clause and leans on internal jargon ('Task Ledger') rather than natural user phrases. It is serviceable but generic within the memory/state niche, risking overlap with similar skills.

Suggestions

Add an explicit trigger clause, e.g. 'Use when the bot starts or ends a task, when asked to recall previous sessions, or before preparing PRs.'

Replace or gloss internal jargon ('Task Ledger') with natural user vocabulary such as 'task history', 'lessons learned', or 'remember prior work'.

Name the concrete artifact (tools/gemini-cli-bot/lessons-learned.md) to sharpen both specificity and distinctiveness from generic memory skills.

Dimension	Reasoning	Score
Specificity	Names the domain and three parallel actions ('maintaining persistent bot memory, synchronizing with previous sessions via the Task Ledger, and preserving decision logs'), but the actions are process-level rather than concrete operations. It falls below anchor 4 because no artifacts, file paths, or concrete operations beyond 'Task Ledger' are named.	3 / 5
Completeness	The 'what' is clear (maintain memory, sync sessions, preserve decision logs) but there is no 'when' clause or equivalent trigger guidance, which caps completeness at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	Some relevant keywords ('memory', 'previous sessions', 'decision logs') exist, but 'Task Ledger' is internal jargon and natural user phrasing ('remember what we did', 'lessons learned', 'recall state') is missing. Not anchor 4 because there is no natural trigger phrasing at all.	3 / 5
Distinctiveness Conflict Risk	'Persistent bot memory' and 'Task Ledger' carve out a niche, but 'memory' and 'session synchronization' overlap significantly with generic memory/state-management skills. No file path or product-specific marker is given to make the niche unambiguous, so it does not reach anchor 4.	3 / 5
	Total	12 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: google-gemini/gemini-cli
Path: tools/gemini-cli-bot/.gemini/skills/memory/SKILL.md
Commit: 361b0bb

Reviewed: 6 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.