Score breakdown
Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.
Why it matters
Useful for AI research and engineering teams that need a deep-research agent that does not silently skip walled sources (the Evidence Record is the headline honesty differentiator — every other agent drops walled pages without saying so): deepcloak is the MIT local-first deep-research agent that escalates individual URLs to a stealth fetch only when a plain request gets a 403, ships a CLI + an MCP
Who should use it
Who should skip it
Skip Mrbaeksang/deepcloak if you cannot isolate its execution environment or audit what data it touches before connecting anything sensitive.
About this signal
Mrbaeksang/deepcloak is tracked by RepoRadar as a local-first deep-research agent in the MIT local-first deep-research agent that reads w section. It was first seen on 2026-06-25 and last updated on 2026-06-25. The current verdict is 'try now' with a Gold tier and moderate setup difficulty. The standout signals for Mrbaeksang/deepcloak are workflow potential (9.5) and novelty (9.0), while momentum (6.0) trails — that balance shapes where it fits best. This page summarizes the evidence RepoRadar has captured from captured source metadata. The score, tier, risk label, and verdict on this page are never influenced by sponsorship, ads, or tips — they reflect only the usefulness, popularity, novelty, momentum, maturity, and evidence signals described in the RepoRadar methodology.
How this item is evaluated
RepoRadar assigned Mrbaeksang/deepcloak a composite score of 8.0 out of 10, placing it in the Gold tier. This score combines weighted sub-signals: usefulness (35%), novelty (18%), momentum (14%), maturity (10%), open-source/build quality (7%), evidence quality (6%), workflow potential (6%), and setup ease (4%). Popularity is tracked separately at 54.0 and never affects the composite score or tier. The risk label of 'medium' reflects inherent user-impacting hazards, not generic novelty. Items with no risk flag may still require normal code review before production use.
Putting this into practice? Read How to vet an AI agent or MCP server before you wire it in for the checklist behind this score.
Risk explanation
**Bot-wall bypass is a capability whose legal posture varies by source and jurisdiction.** The MIT license covers the code, but the act of bypassing a bot wall is governed by the target site's terms of service and by applicable law (CFAA, EU ePrivacy, equivalent regional statutes). The README's local-first framing and Evidence Record are the right design, but the team deploying deepcloak is responsible for confirming the escalation policy matches the target sources' TOS and the team's compliance posture — a strict 'never escalate' policy is the right default for any source that has not been explicitly authorized; **Stealth fetches fingerprint the client and can be detected at the source.** CloakBrowser's stealth surface is good (the maintainer ships an unedited demo of 8 walls bypassed in one pass) but stealth fetches leave traces at the source. The Evidence Record is the audit trail but not a defense — a target that detects a stealth fetch can rate-limit, ban, or escalate. For high-stakes research, point the team's existing research stack at deepcloak's eval methodology (the wall count + per-URL provenance is the durable contribution) rather than the stealth fetch path; **54 stars and an early-stage project — verify the upstream maintenance before adopting.** Deepcloak is at 54 stars with last push 2026-06-05, and the project's value-add is the orchestrator over two upstream projects (LearningCircuit/local-deep-research and CloakBrowser/CloakBrowser). The risk surface is the maintenance cadence of the upstream projects — if either upstream stops being maintained, deepcloak's escalation path degrades. Confirm the upstream commit cadence on both repos before committing to a production research workflow that depends on deepcloak's wall coverage.