checkpoint-mode

Pause for review every N tasks - selective autonomy pattern

Install with Tessl CLI

npx tessl i github:asklokesh/loki-mode --skill checkpoint-mode

What are skills?

Quality

38%

Does it follow best practices?

Impact

Pending

No eval scenarios have been run

Optimize this skill with Tessl

npx tessl skill review --optimize ./agent-skills/checkpoint-mode/SKILL.md

SKILL.md

Review

Evals

Quality

Discovery

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 too abstract and conceptual to be useful for skill selection. It reads more like a pattern name than a functional description, lacking concrete actions, natural trigger terms, and explicit guidance on when to use it. The technical jargon 'selective autonomy pattern' would not match user queries.

Suggestions

Rewrite to describe concrete actions: e.g., 'Implements checkpoints during multi-step workflows, pausing execution to request user confirmation before continuing.'

Add explicit trigger guidance: 'Use when the user asks for review points, checkpoints, approval gates, or wants to verify work before proceeding with batch operations.'

Include natural keywords users would say: 'checkpoint', 'approval', 'confirm before continuing', 'review my work', 'batch processing with review'.

Dimension	Reasoning	Score
Specificity	The description uses vague language like 'pause for review' and 'selective autonomy pattern' without describing concrete actions. It doesn't explain what tasks are being paused, what review entails, or what actions the skill performs.	1 / 3
Completeness	The description fails to clearly answer 'what does this do' beyond a vague concept, and completely lacks any 'when should Claude use it' guidance. There is no 'Use when...' clause or equivalent trigger guidance.	1 / 3
Trigger Term Quality	Contains technical jargon ('selective autonomy pattern') that users would not naturally say. Terms like 'pause for review' and 'N tasks' are abstract and don't match natural user language for requesting this functionality.	1 / 3
Distinctiveness Conflict Risk	The concept of 'selective autonomy' and 'pause for review every N tasks' is somewhat specific to workflow control patterns, but the vague phrasing could overlap with other task management or workflow skills.	2 / 3
	Total	5 / 12 Passed

Implementation

70%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This skill has strong structural organization and clear workflow documentation, making the checkpoint pattern easy to understand. However, it's somewhat verbose with philosophical justification that Claude doesn't need, and the code examples are more illustrative than executable. The core concept is well-communicated but could be tightened.

Suggestions

Remove or significantly condense the Philosophy section - Claude doesn't need persuasion about why to use the pattern

Replace pseudocode with executable code or clearly mark it as a template requiring implementation of helper functions

Consolidate the 'When to Use' section into a brief decision rule rather than extensive bullet lists

Dimension	Reasoning	Score
Conciseness	The skill includes some unnecessary philosophical context and quotes that Claude doesn't need. The comparison table and metrics sections add bulk without being essential for execution. However, the core workflow and configuration sections are reasonably efficient.	2 / 3
Actionability	Provides configuration examples and pseudocode workflow, but the Python code is illustrative rather than executable (references undefined functions like `load_completed_tasks()`, `generate_checkpoint_summary()`). The signal file mechanism is concrete but lacks complete implementation details.	2 / 3
Workflow Clarity	The checkpoint workflow is clearly sequenced with explicit steps: generate summary → create approval signal → wait for approval → resume. The visual diagram and numbered steps make the process unambiguous, with a clear feedback loop for approval.	3 / 3
Progressive Disclosure	Well-organized with clear sections (Overview, When to Use, Configuration, Workflow). References to external files are one level deep and clearly signaled. Content is appropriately structured for a skill of this complexity.	3 / 3
	Total	10 / 12 Passed

Validation

90%

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 — 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

Reviewed: 3 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.