collab-proof

Use when you want to understand what Claude contributed vs what you drove in a session. Triggers on: /collab-proof, session retrospective, ai contribution analysis, collaboration evidence, what did claude do.

Quality

67%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

—

The risk profile of this skill

Fix and improve this skill with Tessl

tessl review fix ./engineering/collab-proof/skills/collab-proof/SKILL.md

Quality

Content

60%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 content is a tightly written, highly actionable 3-layer pipeline with concrete commands, code, and templates, and clear sequencing with honesty-rule validation. Its chief weakness is progressive disclosure: it ignores the four bundled reference files entirely and inlines large blocks that would be better split out.

Suggestions

Link the four references/ files from the body (e.g., under an 'Evidence patterns' or 'References' section) so the bundled material is actually discoverable rather than orphaned.

Move the large fixed HTML structure/CSS specification and the token-collection Python script into dedicated reference or script files, keeping SKILL.md as a lean overview that points to them.

Add an explicit validate/retry feedback loop for the multi-file output writes (e.g., confirm each file was written before proceeding, and a recovery step on failure) to lift workflow clarity.

Dimension	Reasoning	Score
Conciseness	The body is dense and directive with no padding about concepts Claude already knows, but at ~380 lines it is notably long and inlines the full HTML structure spec, a Python token script, and dual DECISIONS.md templates that could be tightened or externalized, so mostly efficient but with room to trim.	3 / 5
Actionability	Provides highly concrete, mostly executable guidance — exact git commands, a complete Python token-collection script, exact CSS hex tokens, and exact HTML class names — but the HTML body itself is only a placeholder skeleton ("... (full HTML with inline CSS)"), so it stops short of fully copy-paste-ready code.	4 / 5
Workflow Clarity	A clearly sequenced 3-layer pipeline (01 signal → 02 classifier → 03 output) with explicit rubrics, pruning rules, exception guards, and honesty rules acting as validation checkpoints; lacks an explicit error-recovery feedback loop for the multi-file writes, so just below 5.	4 / 5
Progressive Disclosure	Four reference files exist in references/ but the body never links to or signals any of them, and content that clearly belongs in separate files (the HTML structure spec, token script, templates) is inlined into a long monolithic body, so structure is minimal and references are effectively orphaned.	2 / 5
	Total	13 / 20 Passed

Description

75%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 strong: it explicitly covers both what the skill does and when to trigger it, uses a slash command plus several natural trigger phrases, and occupies a distinct niche. The main weakness is that the "what" is an abstract single capability rather than a concrete, enumerated action set.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete targets ("session retrospective", "ai contribution analysis", "collaboration evidence", "what did claude do"), but is more trigger-oriented than a comprehensive enumeration of concrete actions, so it sits above 3 but below 5.	4 / 5
Completeness	Explicitly answers both "what" ("understand what Claude contributed vs what you drove") and "when" ("Use when...", "Triggers on:") with concrete trigger phrases, but the "what" is a single abstract capability rather than the multiple concrete operations of the 5-anchor, so 4 not 5.	4 / 5
Trigger Term Quality	Includes a slash-command trigger ("/collab-proof") and natural phrases users would say ("session retrospective", "ai contribution analysis", "what did claude do") with good coverage, but misses some synonyms and extension variations, so above 3 yet not fully comprehensive.	4 / 5
Distinctiveness Conflict Risk	Clear niche (AI collaboration evidence in sessions) with a distinct slash-command trigger and minimal conflict risk, though the abstract "understand what Claude contributed" and "collaboration evidence" phrasing could loosely overlap with general session-summary skills, keeping it just below 5.	4 / 5
	Total	16 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: alirezarezvani/claude-skills
Path: engineering/collab-proof/skills/collab-proof/SKILL.md
Commit: 19392f7

Reviewed: ~6 hours 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.