stevengonsalvez/compound-docs
toolkit/packages/skills/compound-docs/SKILL.md
Internal module for knowledge capture. Called by /reflect for generating structured learning documents with YAML frontmatter and entity sidecars. Not user-invocable — use /reflect or /reflect --knowledge instead.
$ install --global
skillsauthnpx skillsauth add stevengonsalvez/agents-in-a-box compound-docsInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: May 18, 2026, 4:53 AM144.3s4 files scanned
SKILL.md
- name:
- compound-docs
- description:
- |
- version:
- 1.1.0
- author:
- Claude Code Toolkit
- user-invocable:
- false
Internal Module: Called by
/reflectfor knowledge capture. Use/reflectdirectly, or/reflect --knowledgefor knowledge-only capture.
Compound-Docs — Knowledge Note Generator
Purpose
Generate structured learning documents from solved problems. This module handles:
- Context gathering (problem, root cause, solution, files, tags)
- Category auto-detection
- Learning document generation (YAML frontmatter + markdown)
- Entity sidecar generation for GraphRAG indexing
- Saving to
docs/solutions/and promoting to global KB
Context to Gather
- Problem: Error message, observed vs expected behavior
- Root Cause: What actually caused it, why
- Solution: What fixed it, key insight, steps
- Files: Which files were modified or had the bug
- Tags: Technologies involved, searchable keywords
Category Auto-Detection
| Category | Indicators |
|----------|------------|
| build-errors | Compile errors, CI failures, bundling |
| performance-issues | Slowdowns, memory leaks, optimization |
| security-fixes | Vulnerabilities, auth issues, secrets |
| testing-patterns | Test strategies, flaky tests |
| debugging-sessions | Complex investigations |
| architecture-decisions | Design choices, patterns |
| api-integrations | Third-party APIs, SDKs |
| dependency-issues | Package conflicts, upgrades |
| deployment-fixes | Production incidents |
| database-migrations | Schema changes, data fixes |
| ui-patterns | Frontend patterns, CSS |
| tooling-setup | Dev environment, configs |
Learning Document Format
---
title: "[Brief descriptive title]"
category: [auto-detected or specified]
tags: [extracted tags]
symptoms:
- "[Error message or behavior]"
root_cause: "[What actually caused it]"
key_insight: "[THE ONE THING that fixes it]"
created: [today's date]
confidence: [high|medium|low]
language: [if applicable]
framework: [if applicable]
---
## Problem
[Description]
## Solution
[Steps with code examples]
## Context
[Why it happened, how to prevent]
Entity Extraction
For GraphRAG indexing, extract entities and relationships.
See reflect/references/knowledge_format.md for entity types,
relationship types, extraction guidelines, and sidecar format.
Saving
# Project-local
mkdir -p docs/solutions/[category]
# Save: docs/solutions/[category]/[filename].md
# Global promotion (via the reflect CLI from reflect-kb).
# `--force` skips the interactive y/N prompt; content-hash doc_id makes the
# call idempotent so re-runs no-op cleanly.
if command -v reflect >/dev/null 2>&1; then
reflect add docs/solutions/[category]/[filename].md \
--entities docs/solutions/[category]/[filename].entities.yaml \
--force
fi
Quality Checklist
- [ ] Title is descriptive (searchable)
- [ ] Key insight is THE ONE THING (not a summary)
- [ ] Symptoms include actual error messages
- [ ] Tags cover relevant technologies
- [ ] Category is correct
- [ ] Confidence reflects certainty
Related Skills
stevengonsalvez/reflect:cost
documentation
VerifiedTrustedCommunity
Report reflect drain spend over a time window — tokens split by cached (cache_read), uncached writes (cache_creation), and io (input+output), with a $ estimate, grouped by day / outcome / model / transcript. Reads the drainer's cost log and surfaces outlier runs and cache-reuse health (the 41.5M-token failure mode = low cache reuse + high cache writes). Use to answer "what is reflection costing me" for the last day / week.
12SKILL.mdUpdated Jun 2, 2026
stevengonsalvez/ainb-fleet:standup
development
VerifiedTrustedCommunity
Show fleet status — every claude session running on the host, merged across ainb + claude-peers broker + background jobs. Use when you need to enumerate sessions before composing an action, see which sessions have a peer registered (broker-routable) vs tmux-only, check the `summary` of each session, or pipe the list into jq for filtering. Default output: text table. Pass --format json for LLM consumption.
10SKILL.mdUpdated May 31, 2026
stevengonsalvez/ainb-fleet:sequence
testing
VerifiedTrustedCommunity
Ordered multi-step prompts to fleet targets, ack-gated between steps via JSONL assistant-turn-end detection. Use for cycles like disconnect→reconnect→verify, or any flow where step N+1 requires step N to have completed first. The skill BLOCKS until each target's transcript shows the next assistant turn finishing OR per-step timeout fires (default 300s).
10SKILL.mdUpdated May 31, 2026
stevengonsalvez/ainb-fleet:needs
development
VerifiedTrustedCommunity
Center control panel — enumerate every claude session that is blocked waiting on something: a user answer (AskUserQuestion fired), an API error retry, an idle assistant turn-end with no follow-up, or an explicit WAITING: marker. Returns rich JSON with signal kind + context per session. Use this when you've stepped away from the fleet and want one place to see everything that wants your attention and answer it.
10SKILL.mdUpdated May 31, 2026