ReskPoints
An observability and log auditing component for AI agents that records the confidence, parameters, and results of tool calls, automatically masks sensitive fields, and supports log delivery to multiple platforms such as Datadog, Prometheus, and OpenTelemetry.
Timeline (last 90 days)
According to RESK's official documentation, ReskPoints provides enforcement points that place checks at the boundary between the agent runtime and the tools, rather than being a sidecar component that observes traffic after the fact.
ReskPoints provides an audit trail that records authorization decisions and their reconstruction context.
ReskPoints can be installed via pip install reskpoints, and supports installing versions with optional dependencies such as datadog, prometheus, and opentelemetry.
ReskPoints initializes the logging pipeline through AgentLogger(), automatically supplements timestamps, host, environment, and UUID, and allows explicit recording of information such as agent_id, action, probability, params, and result at tool invocation sites.
In reskpoints.yaml, ReskPoints supports enabling masking and declaring sensitive fields (such as api_key, token, secret, password, email, phone), masking sensitive data before logs leave the application.
ReskPoints provides the @log_action decorator, which can automatically record parameters and results for high-risk functions (such as send_email).
ReskPoints only records actions but does not block them; policy enforcement must be done in the Agent code (such as an email domain allowlist); masking relies on field names and regex, and may fail for PII in free text; sampling may miss low-frequency attacks, and sensitive actions should remain fully sampled.
In production environments it is recommended to implement five measures: enable masking for all PII fields, sample 100% of tool calls that access customer data, send logs to a SIEM and set up alerts, use correlation IDs to trace sessions, and regularly audit by replaying logs.jsonl with reskpoints replay.
The scenario in this tutorial is fictional, and Cdiscount does not use, know about, or endorse RESK.
ReskPoints supports sampling rules (such as 1% sampling for heartbeat-type actions, keeping tool_* at 100%) as well as delivery methods such as Console, File, Webhook, Datadog, Prometheus, and OpenTelemetry, with Webhook supporting a signing key to verify the source.
3 more timeline events
All Facts (13)
According to RESK's official documentation, ReskPoints provides enforcement points that place checks at the boundary between the agent runtime and the tools, rather than being a sidecar component that observes traffic after the fact.
50%UnverifiedReskPoints provides an audit trail that records authorization decisions and their reconstruction context.
50%UnverifiedReskPoints can be installed via pip install reskpoints, and supports installing versions with optional dependencies such as datadog, prometheus, and opentelemetry.
50%UnverifiedReskPoints initializes the logging pipeline through AgentLogger(), automatically supplements timestamps, host, environment, and UUID, and allows explicit recording of information such as agent_id, action, probability, params, and result at tool invocation sites.
50%UnverifiedIn reskpoints.yaml, ReskPoints supports enabling masking and declaring sensitive fields (such as api_key, token, secret, password, email, phone), masking sensitive data before logs leave the application.
50%UnverifiedReskPoints provides the @log_action decorator, which can automatically record parameters and results for high-risk functions (such as send_email).
50%UnverifiedReskPoints only records actions but does not block them; policy enforcement must be done in the Agent code (such as an email domain allowlist); masking relies on field names and regex, and may fail for PII in free text; sampling may miss low-frequency attacks, and sensitive actions should remain fully sampled.
50%UnverifiedIn production environments it is recommended to implement five measures: enable masking for all PII fields, sample 100% of tool calls that access customer data, send logs to a SIEM and set up alerts, use correlation IDs to trace sessions, and regularly audit by replaying logs.jsonl with reskpoints replay.
50%UnverifiedThe scenario in this tutorial is fictional, and Cdiscount does not use, know about, or endorse RESK.
50%UnverifiedReskPoints supports sampling rules (such as 1% sampling for heartbeat-type actions, keeping tool_* at 100%) as well as delivery methods such as Console, File, Webhook, Datadog, Prometheus, and OpenTelemetry, with Webhook supporting a signing key to verify the source.
50%UnverifiedReskPoints is positioned as an AI Agent logging component that records the confidence, parameters, and results of each action, automatically masks sensitive fields, and supports multi-platform delivery.
50%UnverifiedReskPoints records the decision flow, with each record carrying the identity, tool name, parameters, and whether the call was allowed or denied.
50%UnverifiedHaving only an allowlist without seeing denial records is like being blind; having only logs without enforcement is nothing but a diary—neither alone is enough.
50%