picoclaw-traffic-guardian

AdvisoryAudited by Static analysis on May 10, 2026.

Overview

No suspicious patterns detected.

Findings (0)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

If a future implementation is enabled, it may be able to see selected gateway traffic, including sensitive prompts, responses, or tokens in transit.

Why it was flagged

The planned implementation would proxy and inspect gateway traffic, including optional HTTPS interception. This is sensitive authority, but the same artifacts restrict it to opt-in, per-process configuration with no automatic system CA installation or default blocking.

Skill content
- lightweight AI gateway HTTP proxy inspection
- optional HTTPS inspection with per-process CA trust
- outbound exfiltration detection
Recommendation

Only enable monitoring for processes you intend to inspect, keep HTTPS trust per-process, and verify that no global proxy or system CA changes are applied automatically.

What this means

Local logs or profile fragments could reveal that sensitive material or suspicious traffic was observed, even if snippets are redacted.

Why it was flagged

The skill plans to persist redacted traffic findings and export posture data for another security profile. This is purpose-aligned, but local logs and profile fragments can still contain security-relevant context and must remain properly redacted and scoped.

Skill content
- redacted local threat logs
- profile export for `picoclaw-security-guardian`
...
6. Emit JSONL findings with redacted snippets.
Recommendation

Keep log/profile output in protected directories, verify redaction behavior before relying on it, and periodically clean up retained findings if they are no longer needed.

What this means

Installing this version mainly installs documentation/spec instructions; actual monitoring behavior would depend on future code not present here.

Why it was flagged

The current package is a specification scaffold with no executable implementation and no declared source repository. That means there is no active code risk in this release, but future proxy code would need separate provenance and security review.

Skill content
Source: unknown
...
No install spec — this is an instruction-only skill.
...
No code files present
Recommendation

Before using any later runtime implementation, confirm the source, review the code, and check that it preserves the documented safety contract.