DeepThink
v1.0.4Manage your personal knowledge, store insights, track tasks, and stay accountable by syncing and updating your DeepThink user data and todos.
⭐ 4· 2.5k·6 current·6 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
SKILL.md clearly describes a connector to an external service (api.deepthink.co) that reads and writes user records, todos, transcripts, and sends messages; that purpose is plausible. However the registry metadata declares no required environment variables or primary credential even though the doc shows Authorization: Bearer dt_live_xxx is required. The skill also has no homepage or source information. The missing declaration of an API key and the absent provenance make the declared capabilities and required artifacts incoherent.
Instruction Scope
Instructions instruct the agent to perform frequent bidirectional syncs, poll live transcripts at each 'heartbeat', proactively contact the user over external messaging channels (e.g., Telegram), and create/patch records and todos. These behaviors go beyond a simple passive lookup: they enable autonomous, ongoing monitoring and outbound communications. The SKILL.md references system-level artifacts (transcripts, messaging channels) and asks the agent to act on them, but does not specify required permissions, credential names, or limits — granting broad discretion to the agent. The document does include a prompt-injection guidance note, which is good, but it does not limit other risky actions.
Install Mechanism
This skill is instruction-only and has no install spec or code files, so nothing will be written to disk by an installer. That lowers supply-chain risk. However, behavior risk derives from the runtime instructions themselves rather than installation.
Credentials
The runtime docs require an API key (Bearer dt_live_xxx) and implicitly require access to the user's messaging channel and transcript sources, but the registry lists no required env vars/config paths and no primary credential. That mismatch is a red flag: the skill will need secrets (DeepThink API key, and likely messaging tokens) but doesn't declare them or explain how they'll be provided or scoped.
Persistence & Privilege
The skill prescribes periodic syncing (every 1–2 days), heartbeat-driven transcript polling, and proactive outbound messaging. Although 'always' is false and autonomous invocation is the platform default, the instructions effectively request ongoing background behavior and external communication — a higher-privilege operational model — without documenting scheduling, limits, or required approvals. Combined with missing credential declarations and unknown provenance, this increases the risk profile.
What to consider before installing
This skill's instructions talk to a live external service (api.deepthink.co) and expect an API key and access to transcripts and messaging channels, but the package metadata declares no credentials and provides no homepage or source. Before installing, ask the publisher for: (1) the official service domain and code/repository or homepage, (2) an explicit list of required environment variables (e.g., DEEPTHINK_API_KEY) and the exact names the skill expects, (3) what messaging channels it will use and what tokens/permissions it requires, and (4) limits on proactive behavior (how often it will poll/send messages and how to disable it). If you can't verify provenance and precise credential requirements, avoid installing or run it in a tightly scoped, auditable environment. If you proceed, only supply the minimal-scoped credentials and test with a throwaway account first.Like a lobster shell, security has layers — review code before you run it.
latestvk97abb0mvyc4ms9sre2780jysx80kvrp
License
MIT-0
Free to use, modify, and redistribute. No attribution required.