conversation-analyzer — agentic threat model

Not certain from the listing — relies on Claude Code's underlying foundation model. Primary threat is indirect prompt injection if malicious payloads exist within the analyzed conversation logs, potentially hijacking the model's analysis output.

Reads local conversation logs to identify patterns. The primary threat is data exposure or exfiltration of sensitive credentials, API keys, or proprietary code that may reside in the local history files being analyzed.

The agent framework is a local productivity plugin. Threats include insecure parsing of conversation logs and potential tool misuse if the 'automation-opportunity surfacing' transitions from passive reporting to active code execution.

Not certain from the listing — runs locally within the user's environment alongside Claude Code. Threats depend on the host environment's security, local file permissions, and whether the plugin runs in a sandboxed process.

Not certain from the listing — no built-in evaluation, logging, or guardrail mechanisms are described. There is a risk of silent failures or biased pattern detection without observability tools.

Not certain from the listing — being an open-source community plugin, it lacks explicit compliance certifications (like SOC2) or built-in access control policies for restricting which local logs can be read.

Operates as a skill/plugin within the Claude Code ecosystem. Threat includes horizontal vulnerability propagation if another compromised local agent or tool manipulates the logs this agent reads.