K Deep Research

Systematic deep research methodology for ANY domain. 7-step workflow with credibility scoring, pattern recognition, adversarial analysis, and iterative deepe...

Security Scan

The skill is a plausible deep-research methodology, but its files and SKILL.md metadata contain contradictions and it instructs reading local files and running long-running/autonomous research patterns—these mismatches and broad scope warrant caution before installing.

Purpose & Capability

The stated purpose (universal deep research methodology) matches the included references and workflow. However the SKILL.md frontmatter declares metadata.openclaw.requires.binsOneOf ["curl","wget"] and metadata.openclaw.alwaysActive: true, while the registry summary lists no required binaries, no env vars, and flags.always=false. That mismatch between declared requirements in the skill body and the registry metadata is an incoherence worth flagging: either the skill expects network tooling and to be always-active (which is significant), or the registry metadata is inaccurate.

Instruction Scope

Runtime instructions tell the agent to load multiple reference files, prioritise SearXNG for sensitive searches, and to consult local files (Obsidian vault, MEMORY.md, TASKS.md, session files referenced elsewhere). The workflow explicitly requires multi-source collection (40–80+ sources) and cron/heartbeat monitoring patterns. Those instructions go beyond a simple passive methodology: they imply repeated filesystem reads and broad network searches. The skill's frontmatter does not declare config paths or any host file access permissions, so the instructions allow the agent to access local data that wasn't declared.

Install Mechanism

No install spec is provided (instruction-only), which is lower-risk from an installation perspective. There is one small helper script (scripts/validate.sh). Because there is no download/install step, nothing new is written to disk by an installer; however the presence of a shell script means someone could execute it to validate or modify the environment — inspect its contents before running. Overall install risk is moderate-to-low if you don't execute included scripts.

Credentials

The skill declares no required environment variables in the registry metadata, yet SKILL.md references functionality that commonly requires credentials or config (notifications via Telegram, connecting to SearXNG instances, reading OpenClaw session files and gateways). The SKILL.md frontmatter also requests curl/wget (binsOneOf) but the registry says none. There is no primaryEnv declared even though the references discuss channels and plugins. This disparity means the skill's expected access surface (network tools, local configs, channel tokens) is not transparently declared.

Persistence & Privilege

Registry flags show always:false (good), but SKILL.md metadata sets alwaysActive:true — a direct contradiction. The skill's references promote cron/heartbeat monitoring and long-lived research pipelines (schedules, notify channels, periodic sweeps). If enabled with autonomous invocation and given access to cron/heartbeat or gateway channels, the skill could be used for ongoing background monitoring. Autonomous invocation alone is normal, but the mismatch about 'always active' and the explicit guidance to integrate into cron/heartbeat are reasons to restrict automated scheduling and to avoid granting broad persistent privileges without review.

What to consider before installing

This skill appears to be a genuine research methodology pack, but several things do not add up and deserve attention before enabling it: - Metadata mismatch: SKILL.md claims it needs curl/wget and marks itself alwaysActive=true, but the registry metadata lists no binaries and always:false. Treat the SKILL.md frontmatter as authoritative until corrected — ask the author to reconcile or update registry metadata. - Local file access: The workflow explicitly expects the agent to consult local stores (Obsidian vault, MEMORY.md, TASKS.md, OpenClaw session files). If you install it, ensure the agent is NOT granted access to sensitive directories (.ssh, .env, home) unless you intend that. Prefer sandboxed testing. - Autonomous monitoring: The references encourage cron/heartbeat integration and automated monitoring. Do not wire this skill into scheduled jobs, gateway channels, or notification hooks (Telegram, etc.) until you’ve inspected and limited what it can access and send. - Inspect scripts: Open scripts/validate.sh and any other scripts; do not run them as root or on a production host until you review their behavior. - Least privilege test: Install in a sandbox workspace with minimal permissions; run a trivial research prompt and observe what files/tools the skill tries to use. If it attempts to read host-level paths or request tokens not declared in metadata, stop. - Ask the author: Request clarification about (1) whether the skill must be alwaysActive, (2) what binaries and external endpoints it expects (SearXNG instance URL?), and (3) any notification/channel integrations and required env vars. Require the registry metadata to be corrected before granting broader access. If you want to proceed safely: load it into an isolated OpenClaw workspace, deny network or host file access initially, review behavior, and only then selectively grant limited permissions for intended use.

K Deep Research v2.0

Universal research methodology for any domain, any topic, any complexity level. Optimized for OpenClaw autonomous agents AND Claude.ai project workflows.

⚠️ CRITICAL: Load Before Researching

When research is requested, you MUST:

1. Read this SKILL.md (you're doing it now — good)
2. Load references/sourcing-strategies.md — WHERE and HOW to search
3. Load domain-relevant references as needed (see Reference Map below)
4. Execute the 7-step workflow
5. Output as Obsidian-ready .md file (YAML frontmatter mandatory)

DO NOT skip this skill and jump to web search. Methodology > raw queries.

Core Research Workflow

Execute in sequence for every investigation:

1. CONTEXT CHECK    → Existing knowledge base / prior research
2. QUERY ELABORATION → Expand scope, plan search strategy
3. MULTI-SOURCE      → Gather from diverse sources (40-80+ for deep)
4. PATTERN ANALYSIS  → Cross-domain recognition, temporal/actor/info flow
5. CREDIBILITY SCORE → 0-10 scale on ALL sources, merit-based
6. SYNTHESIS         → Compile findings preserving contradictions
7. OUTPUT            → Obsidian .md with YAML frontmatter

Research Principles

Institutional Skepticism: Official narratives = data points, not truth claims. Merit-Based Sources: All sources start equal. Evaluate on internal consistency, specificity, predictive accuracy, corroboration potential, incentive analysis, technical coherence. Peer review is not a truth guarantee; institutional rejection is not falsification. Pattern Recognition: Temporal clustering, actor coordination, information flow, anomaly correlation, historical precedent, narrative consistency. Epistemic Humility: Absence of evidence ≠ evidence of absence. BUT systematic patterns of absence ARE informative. Physics First: Technical feasibility analysis before accepting exotic claims. Adversarial Analysis: Cui bono? Suppression signatures? Inversion test (what if the "debunking" is the disinformation)?

Tool Selection Strategy

SearXNG (PRIMARY for sensitive/adversarial research):

• Zero telemetry, aggregates across engines
• Use for: institutional analysis, suppression tracking, contested topics
• Fallback: built-in web_search when SearXNG unavailable

Web Search (general research):

• Current events, academic papers, community discussions
• Non-sensitive technical topics

Context7 MCP (technical documentation):

• Code libraries, frameworks, APIs, SDKs
• Coverage: 30k+ snippets across dev ecosystem
• NOT for: consciousness, legal, historical, institutional topics

Filesystem (existing knowledge):

• Obsidian vault (4000+ files)
• Prior investigation notes, timelines, frameworks

Decision Tree:

Sensitive/adversarial topic?  → SearXNG first
Code/framework/API docs?      → Context7 first
Existing research available?  → Filesystem first
General research?             → Web search
Always:                       → Multi-source triangulate

Source Credibility Scale (Merit-Based)

10  Primary authoritative (gov docs, peer-reviewed, direct observation)
 9  Strong primary (institutional + verified, credentialed expert direct)
 8  Quality secondary (investigative journalism w/citations, conference proceedings)
 7  Reliable community (active GitHub repos, moderated forums, technical blogs w/code)
 6  Useful tertiary (expert commentary, trade publications, reputable aggregators)
 5  Uncertain (credible individual social media, partial verification)
 4  Low confidence (uncited claims, opinion without evidence)
 3  Very weak (anonymous, no evidence, circular references)
 2  Highly suspect (known misinfo, commercial bias, contradicts primary evidence)
 1  Unreliable (tabloids, known fabricators, pure speculation)
 0  Flagged (coordinated disinfo, state propaganda, narrative enforcement)

CRITICAL: Score reflects evaluated merit, NOT source prestige. A forum post with technical depth and internal logic may outrank mainstream article amplifying official statements.

Output Format (Default: Obsidian .md)

Every report gets YAML frontmatter:

---
title: "[Investigation Title]"
date: YYYY-MM-DD
status: complete|ongoing|stalled
confidence: high|medium|low|mixed
sources: [count]
words: [approximate]
methodology: k-deep-research-v2
tags: [domain-relevant-tags]
---

Report structure scales to complexity:

• Executive synthesis (quick reference, NOT replacement for depth)
• Full hierarchical body (Parts → Sections → Subsections)
• Every claim supported, every thread followed
• Technical appendices where applicable
• Comprehensive sourcing with credibility scores
• Unanswered questions and future investigation vectors

LENGTH IS A FEATURE. 10,000+ words exhausting a topic = SUCCESS. 2,000 words hitting highlights = FAILURE.

Confidence Levels

State for ALL key conclusions:

• HIGH: Multiple independent sources, physical evidence, internally consistent
• MEDIUM: Credible sources but limited corroboration, or logical inference from HIGH data
• LOW: Single source, circumstantial, or pattern extrapolation
• SPECULATIVE: Hypothesis consistent with data but unverified — mark clearly

Dead End Protocol

When investigation stalls:

1. Document what was searched and what returned nothing
2. Distinguish "no evidence found" vs "evidence likely inaccessible/suppressed"
3. Note absence patterns — systematic gaps ARE data
4. Flag for future: "Revisit if [condition] changes"
5. Don't spin wheels — acknowledge, document, move on

Tool Failure Protocol

When tools fail (rate limits, paywalls, MCP errors):

1. Note failure and what was attempted
2. Route around: alternative sources, cached versions, archive.org, adjacent queries
3. Don't silently omit — "Attempted X, blocked by Y, pivoted to Z"
4. Pattern of access failures may itself be informative

Reference Files — Load As Needed

Always Load First

• references/sourcing-strategies.md — WHERE to find info, HOW to construct queries, multi-source triangulation, when to stop searching

Load By Domain

• references/research-frameworks.md — Multi-layer analysis (5 layers), credibility evaluation, information control detection, triangulation methodology, iterative deepening, quality checklist
• references/output-templates.md — Format examples, selection guide, adaptive guidelines
• references/openclaw-architecture.md — OpenClaw Gateway/Agent Runtime architecture, heartbeat daemon, memory systems, model failover, sub-agents, Lobster workflows, session management, tool policy
• references/openclaw-skill-authoring.md — SKILL.md format, YAML frontmatter spec, three-tier loading, reference file patterns, ClawHub registry, security model, testing, publishing
• references/autonomy-patterns.md — Proactive agent patterns, heartbeat vs cron, memory persistence, compaction survival, task registries, workflow orchestration, degradation monitoring, multi-agent coordination
• references/adversarial-analysis.md — Suppression detection, institutional behavior, narrative flow analysis, information archaeology, inversion testing, incentive mapping

Loading Strategy

Research request arrives →
  1. ALWAYS: sourcing-strategies.md
  2. IF complex multi-domain: research-frameworks.md
  3. IF OpenClaw/agent topic: openclaw-architecture.md + autonomy-patterns.md
  4. IF building skills: openclaw-skill-authoring.md
  5. IF institutional/suppression angle: adversarial-analysis.md
  6. IF custom output needed: output-templates.md

OpenClaw Autonomy Integration

When this skill runs inside OpenClaw:

• Heartbeat context: Can be triggered by heartbeat to check research queues
• Cron scheduling: Schedule recurring research sweeps on monitored topics
• Memory persistence: Write research state to MEMORY.md / memory plugin
• Sub-agent delegation: Spawn focused sub-agents for parallel source gathering
• Task registry: Read TASKS.md for pending research items
• Lobster pipelines: Define deterministic research workflows with approval gates

Quality Checklist (Before Completing)

• Loaded sourcing-strategies.md before searching
• Used appropriate tools for domain (SearXNG/Context7/web/filesystem)
• Scored ALL sources for credibility (0-10)
• Documented contradictions explicitly
• Checked for information control patterns (if applicable)
• Applied cross-domain pattern recognition
• Preserved uncertainty where warranted
• YAML frontmatter present with all fields
• Listed next investigation priorities
• Complete source bibliography with scores
• No forced conclusions — evidence speaks

Remember

This methodology is universal. What changes: domain-specific sources and authorities. What stays constant: credibility scoring, pattern recognition, triangulation, epistemic humility.

When K asks a question, the answer is a complete investigation, not a response.