context-compactor

Automatic context summarization for long-running sessions. Use when context is approaching limits, summarizing completed work, preserving critical information, or managing token budgets.

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/context-compactor/SKILL.md

Quality

Content

65%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 well-organized, lean, and gives concrete executable code, but it is weakened by a destructive compaction operation that lacks a concrete validation/feedback loop and by a broken reference to a non-existent TOKEN-BUDGETS.md file.

Suggestions

Add a concrete validation feedback loop for compaction, e.g. after `compact(context)` run a check that critical items (architectural decisions, unresolved bugs) survive and re-compact or back off if not — this satisfies the destructive/batch validation requirement and lifts workflow_clarity above 3.

Remove or create the missing `references/TOKEN-BUDGETS.md` referenced in the References section, since broken links break progressive-disclosure navigation.

Make the Quick Start self-contained: define or stub the `context` variable and replace the redundant second code block with a runnable example that shows preserved vs discarded output.

Dimension	Reasoning	Score
Conciseness	The body is lean with terse tables and code rather than padded explanations of concepts Claude already knows, but the ASCII priority box could be tightened into prose, placing it noticeably above anchor 3 yet short of the every-token-earns-its-place anchor 5.	4 / 5
Actionability	Quick Start gives concrete, mostly-executable Python against the real ContextCompactor API, but `context`/`limit` are undefined, the second snippet repeats the first, and the preserves/discards note is a comment rather than an executable pattern, so anchor 4 fits better than the copy-paste-ready anchor 5.	4 / 5
Workflow Clarity	A 5-step Compaction Process with a Validate step is sequenced, but because compaction permanently discards context the validation must be concrete with a fix-and-retry feedback loop; "Ensure compacted context is coherent" is only an implicit checkpoint, so per the destructive/batch cap this cannot exceed anchor 3.	3 / 5
Progressive Disclosure	The body is an overview with one-level-deep, clearly signaled references and four real scripts, but `references/TOKEN-BUDGETS.md` is referenced and does not exist — a broken navigation link is a real organization failure that pulls this below the minor-gaps anchor 4 toward anchor 3.	3 / 5
	Total	14 / 20 Passed

Description

78%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 concise, well-structured description that clearly answers both what the skill does and when to use it, with good natural trigger terms. It is slightly held back by generic rather than fully specific action descriptions and minor overlap risk with adjacent context-management skills.

Dimension	Reasoning	Score
Specificity	Names the domain plus a few concrete actions ("summarizing", "preserving critical information", "managing token budgets") but the actions are generic rather than comprehensive, matching anchor 3.	3 / 5
Completeness	Explicitly answers both what ("Automatic context summarization for long-running sessions") and when ("Use when context is approaching limits, summarizing completed work...") with concrete trigger phrases, matching anchor 5.	5 / 5
Trigger Term Quality	Good natural-term coverage ("context is approaching limits", "summarizing completed work", "preserving critical information", "managing token budgets", "long-running sessions") with a few common synonyms missing, fitting anchor 4 rather than the comprehensive anchor 5.	4 / 5
Distinctiveness Conflict Risk	Context compaction / token-budget niche is mostly distinct with only minor overlap risk against memory or session-management skills, fitting anchor 4 rather than the minimal-conflict anchor 5.	4 / 5
	Total	16 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 1 missing	Warning

	Total	14 / 16 Passed

Repository: fernandezbaptiste/Skrillz
Path: skills/context-compactor/SKILL.md
Commit: 93ed392

Reviewed: 1 day 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.