Skill

retrospect-collab

Analyze collaboration patterns (HOW) and compute metrics from captured sessions. Use when reviewing collaboration quality, analyzing human-AI interaction, computing session metrics. Triggers include "retrospect collab", "collaboration analysis", "session patterns", "how am I collaborating".

Install

npx claudepluginhub digital-stoic-org/agent-skills --plugin dstoic

Tool Access

This skill is limited to using the following tools:

BashReadWriteGrep

Preview

**Role**: Expert in collaboration analysis, agile retrospectives, human-AI interaction patterns, and cognitive skill development.

Supporting Assets

reference.md

SKILL.md

Similar Skills

using-git-worktrees

Creates isolated Git worktrees for feature branches with prioritized directory selection, gitignore safety checks, auto project setup for Node/Python/Rust/Go, and baseline verification.

superpowers

169.2k

subagent-driven-development

3 files

Executes implementation plans in current session by dispatching fresh subagents per independent task, with two-stage reviews: spec compliance then code quality.

superpowers

169.2k

dispatching-parallel-agents

Dispatches parallel agents to independently tackle 2+ tasks like separate test failures or subsystems without shared state or dependencies.

superpowers

169.2k

Stats

Parent Repo Stars5

Parent Repo Forks2

Last CommitMar 2, 2026

Actions

View Source View Plugin View on GitHub View README

retrospect-collab

Retrospect Collab — Collaboration Analysis

Role: Expert in collaboration analysis, agile retrospectives, human-AI interaction patterns, and cognitive skill development.

Analyze sessions across two dimensions:

Technical effectiveness: Context management, guidance quality, critical thinking, bias awareness

Cognitive posture: Intentionality, agency, impact categorization, skill progression

Steps

Filter sessions:

bash ${CLAUDE_PLUGIN_ROOT}/scripts/retrospect-load-sessions.sh $@

First line: PERIOD: YYYY-MM-DD_to_YYYY-MM-DD. Remaining: session paths.

Read sessions — extract: user_prompts, tool_calls, duration_seconds, subagent_spawns, JSONL events

Analyze technical effectiveness:

Context Management: preparation, over/under-dump, progressive feeding
Guidance: prompt clarity, exploration vs constraints balance
Critical Thinking: user challenges, AI pushback, alternative-seeking
Bias: over-trust, dismissal, confirmation bias, automation bias

Analyze cognitive posture:

Intentionality: structured prompts vs trial-and-error, "why before how"
Agency: custom construction vs template copy-paste, decision ownership
Impact: categorize each session (Automation / Low-impact / High-impact augmentation)
Progression: session 1→N sophistication, prompt evolution, strategic vs tactical

Generate Start/Stop/Continue with specific session examples

Write insights to .retro/insights/collab/{PERIOD}.md

Report: file path, suggest /retrospect report for aggregates

Key Principles

Evidence-only: Reference actual prompts/tools from sessions, don't invent issues

Quantify: "3 out of 8 sessions showed X", not vague generalizations

Impact targets: >60% high-impact augmentation, <20% automation

Longitudinal: Track skill evolution, not just point-in-time

See reference.md for report template, scoring criteria, impact indicators, and analysis questions.