session-management
Context preservation, tiered summarization, resumability
23
13%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/session-management/SKILL.mdQuality
Discovery
0%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description is extremely weak across all dimensions. It consists of three abstract noun phrases with no verbs, no concrete actions, no explicit trigger guidance, and no indication of when the skill should be selected. It reads more like a feature list tagline than a functional skill description.
Suggestions
Rewrite using active verbs describing concrete actions, e.g., 'Summarizes long conversations into tiered levels of detail, preserves key context across sessions, and enables resuming interrupted workflows.'
Add an explicit 'Use when...' clause with natural trigger terms, e.g., 'Use when the user asks to summarize a conversation, resume a previous session, catch up on context, or pick up where they left off.'
Clarify the specific domain or scenario this skill applies to (e.g., long chat sessions, multi-turn projects, document review) to reduce conflict risk with other skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description uses abstract nouns ('context preservation', 'tiered summarization', 'resumability') without describing any concrete actions. There are no verbs indicating what the skill actually does. | 1 / 3 |
Completeness | The description fails to answer both 'what does this do' and 'when should Claude use it'. There is no 'Use when...' clause and the 'what' is only vaguely implied through abstract noun phrases. | 1 / 3 |
Trigger Term Quality | The terms used are technical jargon ('tiered summarization', 'resumability', 'context preservation') that users would almost never naturally say when requesting help. Users would more likely say things like 'summarize our conversation', 'pick up where we left off', or 'remember what we discussed'. | 1 / 3 |
Distinctiveness Conflict Risk | The description is so vague that it's unclear what domain or task it applies to. 'Context preservation' and 'summarization' could overlap with many different skills related to conversations, documents, or data processing. | 1 / 3 |
Total | 4 / 12 Passed |
Implementation
27%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill has a solid conceptual framework for session management with useful tiered checkpointing, but suffers severely from verbosity and repetition — the same decision tree appears in at least 4 different forms. The extensive example templates (JWT auth, Drizzle ORM) pad the content significantly without adding instructional value. The monolithic structure with no bundle files means Claude must process ~400+ lines of mostly redundant content every time.
Suggestions
Consolidate the checkpoint decision logic to a single authoritative location (the decision heuristic diagram) and remove the 3-4 redundant restatements in CLAUDE.md rules, enforcement, and quick reference sections.
Extract the lengthy template examples (current-state.md, decisions.md, code-landmarks.md, archive template) into separate bundle files and reference them from the main skill with one-line descriptions.
Remove the illustrative example data (JWT/RS256, Drizzle ORM) from templates — use minimal placeholders like `[decision title]` and `[option 1]` instead, since Claude can generate contextually appropriate content.
Cut the anti-patterns section entirely (it restates the positive instructions as negatives) and trim the enforcement mechanisms to just the self-check questions, reducing overall length by ~40%.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~400+ lines. Contains extensive template examples that are largely illustrative rather than instructive, explains concepts Claude already understands (what session management is, why context loss is bad), and repeats the same checkpoint decision logic in at least 4 different places (tiered rules, decision heuristic, CLAUDE.md rules, enforcement self-check, quick reference). The anti-patterns section states obvious negations of the instructions already given. | 1 / 3 |
Actionability | Provides concrete file templates and directory structures which are useful, but much of the content is example/placeholder data (JWT auth examples, Drizzle ORM decisions) rather than executable guidance. The bash aliases are actionable but minor. The core skill is a process/workflow skill, and the templates are reasonably specific, but the repeated decision heuristics and enforcement mechanisms are more aspirational than directly executable. | 2 / 3 |
Workflow Clarity | The tiered summarization system (Tier 1/2/3) provides a clear multi-step framework with triggers and actions. However, the same workflow is restated in multiple places (tiered rules, decision heuristic diagram, CLAUDE.md section, enforcement section, quick reference) creating confusion about which is authoritative. There are no validation/verification steps to confirm checkpoints were done correctly — the 'enforcement mechanisms' are just reminders to do the work, not actual verification. | 2 / 3 |
Progressive Disclosure | No bundle files are provided, yet this is a monolithic ~400-line document that would greatly benefit from splitting. The decision log template, code landmarks template, archive template, and CLAUDE.md integration section could each be separate referenced files. Everything is inline, making the skill a wall of text that's hard to scan and navigate. | 1 / 3 |
Total | 6 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (529 lines); consider splitting into references/ and linking | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- Repository
- alinaqi/claude-bootstrap
- Commit
7e5f7a2
- 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.