compact
Compact current session memory into structured text for session recovery. Supports custom descriptions and tagging.
48
36%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Risky
Do not use without reviewing
Optimize this skill with Tessl
npx tessl skill review --optimize ./.codex/skills/memory-compact/SKILL.mdQuality
Discovery
32%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 conveys the core purpose of compacting session memory but lacks explicit trigger guidance ('Use when...'), which significantly hurts completeness. The trigger terms are somewhat relevant but miss common natural language variations users might employ. Adding explicit usage triggers and more concrete action details would substantially improve skill selection accuracy.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when the user asks to save session state, compress memory, checkpoint the conversation, or prepare for session handoff.'
Include more natural trigger terms users might say, such as 'save context', 'compress memory', 'checkpoint', 'persist conversation', or 'session handoff'.
Expand the concrete actions listed—e.g., 'Summarizes conversation history, extracts key decisions and context, and outputs tagged structured text for resuming sessions later.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (session memory) and some actions (compact, session recovery, custom descriptions, tagging), but doesn't list multiple concrete actions in detail—e.g., what 'compact' means specifically or what structured text looks like. | 2 / 3 |
Completeness | Describes what the skill does (compact session memory into structured text) but has no explicit 'Use when...' clause or equivalent trigger guidance, which per the rubric caps completeness at 2, and the 'when' is entirely missing, placing it at 1. | 1 / 3 |
Trigger Term Quality | Includes some relevant terms like 'session memory', 'session recovery', and 'tagging', but misses natural user phrases like 'save context', 'remember conversation', 'compress memory', 'checkpoint', or 'persist state'. | 2 / 3 |
Distinctiveness Conflict Risk | The concept of 'session memory compaction' is fairly niche, but the terms 'memory' and 'tagging' could overlap with other memory management or note-taking skills. It's somewhat specific but not clearly delineated. | 2 / 3 |
Total | 7 / 12 Passed |
Implementation
39%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill has a well-defined workflow with clear sequencing and validation, but is severely bloated with redundant explanations, illustrative pseudocode, and inline reference material that should be externalized. The core actionable content (structured output template + MCP save call) could be conveyed in roughly 1/4 of the current length. Repeated emphasis on preserving plans verbatim and using absolute paths, while important, is stated far too many times.
Suggestions
Reduce content by at least 60%: remove the field definitions table (duplicates the template), collapse the JavaScript pseudocode into brief prose instructions, and eliminate repeated admonitions about verbatim plans and absolute paths.
Extract the plan detection priority logic, path resolution rules, and reference file categories into a separate REFERENCE.md file, keeping only a brief summary and link in the main skill.
Replace the illustrative JavaScript pseudocode (extractTodosFromConversation, inferPlanFromDiscussion) with brief prose descriptions of what to look for, since these aren't executable functions.
Consolidate the structured output format and the generation code into a single section—currently the template appears twice (once as markdown, once inside JavaScript string interpolation).
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~300+ lines. The field definitions table duplicates information already shown in the structured output format. The JavaScript code examples for session analysis, plan detection, and text generation are implementation details Claude doesn't need spelled out this explicitly. Multiple sections repeat the same concepts (e.g., 'preserve complete plan verbatim' is stated 5+ times). | 1 / 3 |
Actionability | Provides concrete structured output format and MCP tool calls, which is good. However, much of the JavaScript is pseudocode/illustrative rather than truly executable (e.g., `extractTodosFromConversation()`, `inferPlanFromDiscussion()` are undefined functions). The MCP call example is actionable, but the surrounding code is more conceptual scaffolding than copy-paste ready. | 2 / 3 |
Workflow Clarity | The 4-step execution flow (Analyze → Generate → Import → Report) is clearly sequenced with explicit outputs at each stage. The quality checklist serves as a validation checkpoint before saving. Plan detection has a clear priority order. The workflow is well-structured for a multi-step process. | 3 / 3 |
Progressive Disclosure | This is a monolithic wall of text with no references to external files. The field definitions table, path resolution rules, reference file categories, and plan detection priority details could all be split into separate reference documents. Everything is inlined, making the skill overwhelming to parse. | 1 / 3 |
Total | 7 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- catlog22/Claude-Code-Workflow
- Commit
0f8e801
- 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.