agentic-jujutsu

Quantum-resistant, self-learning version control for AI agents with ReasoningBank intelligence and multi-agent coordination

1.15x

Quality

39%

Does it follow best practices?

Impact

88%

1.15x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/agentic-jujutsu/SKILL.md

Quality

Content

50%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 rich with executable code and a thorough API reference, making it actionable, but it is over-long and repetitive with inlined content that belongs in separate reference files. Workflow sequencing is reasonable yet missing explicit validation checkpoints for destructive operations.

Suggestions

Split the API reference tables and the four advanced use cases into separate reference files (e.g. REFERENCE.md, EXAMPLES.md) and link to them from SKILL.md to reduce the monolith.

Collapse the redundant multi-agent examples into one canonical example and remove the marketing performance table.

Add explicit validation checkpoints for destructive ops, e.g. after merge/rebase run `jj status` and abort/retry on conflict before finalizing the trajectory.

Dimension	Reasoning	Score
Conciseness	The 640-line body is noticeably padded: multi-agent coordination is demonstrated four times (core capability 4, Use Case 2, Best Practice 3, Example 2), and marketing claims ('23x faster than Git', '87% success rate', a performance table) add tokens without instructive value.	2 / 5
Actionability	Most guidance is concrete and copy-paste ready — real method calls, an API reference table, and an install command — with only minor gaps where helper functions like executeOperation/verifyDeployment are undefined placeholders.	4 / 5
Workflow Clarity	The trajectory sequence (startTrajectory → operations → addToTrajectory → finalizeTrajectory) is clear and there is an input-validation section, but destructive/batch operations (merge, rebase, deploy, enableEncryption) lack explicit verify-before-proceed checkpoints, capping clarity at 3.	3 / 5
Progressive Disclosure	Section headers give some structure, but the SKILL.md is a ~640-line monolith with the full API reference, four use cases, and examples all inlined; no bundle files exist (references/, scripts/, assets/ absent) and the only references point to external package docs.	3 / 5
	Total	12 / 20 Passed

Description

28%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 is buzzword-heavy and reads like marketing copy: it states a vague purpose but gives no concrete actions and no trigger guidance for when to use the skill. It would rarely match a user's natural request phrasing.

Suggestions

Rewrite in third person with concrete actions, e.g. 'Commits, branches, and rebases via a jj wrapper that records agent trajectories and suggests past successful operation sequences.'

Add an explicit trigger clause: 'Use when multiple AI agents need concurrent, lock-free version control or when reusing learned operation patterns for repetitive tasks.'

Drop unverifiable marketing terms ('Quantum-resistant', 'ReasoningBank intelligence') from the description; move them into the body if needed.

Dimension	Reasoning	Score
Specificity	Names the domain ('version control for AI agents') but the stated actions are generic buzzwords ('self-learning', 'multi-agent coordination', 'ReasoningBank intelligence') rather than concrete operations like commit/branch/merge.	2 / 5
Completeness	It offers a vague 'what' and entirely omits any 'when'/'Use when...' trigger guidance, matching the anchor for a vague what with no when; the missing trigger clause caps completeness below the midpoint.	2 / 5
Trigger Term Quality	Only 'version control' and 'AI agents' are plausible user phrases; 'Quantum-resistant' and 'ReasoningBank intelligence' are jargon and marketing buzzwords users would not naturally say, and common synonyms/triggers are absent.	2 / 5
Distinctiveness Conflict Risk	'Version control for AI agents' carves a somewhat specific niche, but the broad buzzword framing ('self-learning', 'multi-agent coordination') leaves overlap risk with general VCS or agent-orchestration skills.	3 / 5
	Total	9 / 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
skill_md_line_count	SKILL.md is long (649 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: ruvnet/agentic-flow
Path: .claude/skills/agentic-jujutsu/SKILL.md
Commit: e993605

Reviewed: 9 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.