The description lists numerous specific concrete actions: 5-stage recon pipeline, asset-graph discipline, severity rubric, confidence upgrade workflows, identity-fabric mapping, breach×identity correlation, detectability tagging, WAF/CDN bypass, vulnerability prioritization, phishing infrastructure planning, bug bounty submission, and client deliverable templates. This is highly specific and comprehensive.

The description clearly answers both 'what' (comprehensive OSINT methodology covering specific stages and techniques) and 'when' with an explicit 'Use when...' clause covering planning/executing reconnaissance, mapping attack surfaces, investigating entities, and producing deliverables.

While it includes some natural terms like 'reconnaissance', 'attack surface', 'red-team', 'bug bounty', and 'phishing', many terms are specialized jargon (e.g., 'asset-graph discipline', 'confidence upgrade workflows', 'identity-fabric mapping', 'breach×identity correlation') that users are unlikely to naturally say. Common user terms like 'OSINT', 'recon', and 'external assessment' are present but the heavy jargon reduces natural keyword coverage.

The description carves out a very clear niche: external red-team OSINT reconnaissance with specific methodologies. The combination of OSINT, red-team operations, authorized assessments, and the detailed technique list makes it highly unlikely to conflict with other skills.

The skill is extensive (~480 lines) and covers a lot of ground efficiently for its scope, but includes some content Claude already knows (e.g., explaining what UTC is, basic concepts like 'VPNs and residential proxies exist'). The anti-patterns section restates things already covered elsewhere. Some tables could be tighter. However, given the breadth of the methodology, most content earns its place.

The skill provides structured frameworks, severity rubrics, confidence upgrade workflows, and clear decision trees — all highly actionable at the methodology level. However, it deliberately defers all concrete implementation (curl commands, regexes, wordlists, probe paths) to a companion skill 'offensive-osint' that is not included in the bundle. The skill tells you *what* to do but not *how* to do it for most technical steps, making it incomplete on its own.

The 5-stage pipeline is clearly sequenced with explicit ordering, time budgets per engagement size, priority ordering within stages, abort conditions, and a detailed detection-aware back-off ladder (§6.4) that serves as a validation/feedback loop for risky operations. Confidence upgrade workflows provide explicit checkpoints for promoting findings. The scope check acts as a gate before any work begins.

The skill references a companion skill 'offensive-osint' with specific section numbers (§13, §16.1, §22, §29.2, etc.) for implementation details, which is good progressive disclosure design. However, no bundle files are provided, meaning all those references are unverifiable and potentially broken. The SKILL.md itself is a single large document that could benefit from splitting (e.g., severity rubric, asset taxonomy, and deliverable templates as separate files).