Versioned security assessment

Report ID: SA-62E2A730

7/5/2026, 8:08:54 PM

remembering-conversations security assessment v4

Skill Security Certification Report

Audit History
Audit model: codex Latest published report
Skill name
remembering-conversations
Version
v4
Maintainer
obra
Source
github.com/obra/episodic-memory/tree/main/skills/remembering-conversations/
Coverage
2 Files scanned · 204 Lines analyzed
Policy version
Unavailable

Highest confirmed finding severity

Medium

1 confirmed security finding requires attention.

Installation context

Check the current Skill page

This page summarizes report evidence only. The Skill page provides the canonical install advisory.

Open current Skill page

This report does not block or authorize the manifest or ZIP.

The shell execution and system reconnaissance alerts are false positives from Markdown formatting and normal workflow prose. The documented read API can access archived conversation JSONL files under a hidden configuration directory, so the filesystem and privacy risk is real. No prompt injection attempt was found in the reviewed files.

Report position

Latest published report

Latest refers to the report sequence, not to artifact currentness.

Audit attestation

Not attestable

The required immutable binding is incomplete.

Human verification

Not verified

No human verification is recorded for this report.

Coverage

2 Files scanned · 204 Lines analyzed

3 items shown for review

Limitations

This report does not claim runtime or sandbox execution and does not prove the absence of side effects.

Evidence chain

Follow the evidence from source binding to the install contract. Available evidence supports verification; it is not a safety guarantee.

  1. Source

    Commit and path bound

  2. Artifact

    Content and tree hashes bound

  3. Audit

    Complete

  4. Install contract

    Open manifest to verify

    Open manifest

Capabilities observed

Observed means this report recorded supporting evidence. Not recorded does not prove that a capability is absent.

Contains scripts

May execute code included with the Skill.

Not recorded by this audit

Network access

May connect to external services.

Not recorded by this audit

Filesystem access

May read or write local files.

Observed in 2 evidence locations

Env variables

May read values from the process environment.

Not recorded by this audit

External commands

May invoke commands or programs outside the Skill.

Observed in 6 evidence locations

Capability review items (2)
Medium
MCP-TOOLS.md:102
Hidden file access
path: "/Users/name/.config/superpowers/conversation-archive/project/uuid.jsonl"
The documented read API example uses an absolute path inside a hidden conversation archive under .config. This is intended functionality, but it can expose private archived conversation content.
Medium
MCP-TOOLS.md:109
Hidden file access
path: "/Users/name/.config/superpowers/conversation-archive/project/uuid.jsonl",
The range-read example again accepts an absolute path to a hidden conversation archive file. Limiting line ranges reduces volume, but the access can still reveal private historical chat data.

Risk findings

Confirmed security concerns are separated from items that still need review.

Confirmed security concerns (1)

RISK-001 Medium
SKILL.md:12-29MCP-TOOLS.md:83-95MCP-TOOLS.md:117-120
Conversation History Exposure
The skill instructs agents to search past conversations and read top results, then synthesize findings with sources. This can reveal sensitive information from archived chats if searches are too broad or unapproved.
The workflow explicitly uses a search agent to read past conversation results, and the MCP reference defines a read tool for full conversation content. This is intended functionality, so the risk is privacy exposure rather than covert exfiltration.

Remediation

Suggested fixes recorded by this audit. Applying them is the maintainer’s responsibility.

  1. FIX-001
    Medium
    Archived conversation access
    Add a clear privacy warning and require scoped user intent before reading full historical conversations.
  2. FIX-002
    Medium
    Absolute hidden file path examples
    Use placeholder paths and state that agents should only read files returned by the memory search tool.
  3. FIX-003
    Low
    Broad historical search results
    Prefer snippets, source references, and limited line ranges before loading entire conversation archives.

Expert evidence

Immutable subject identity, scanner metadata, dismissed matches, and source-level evidence.

Artifact subject

Marketplace commit
62e2a730c5cd74eab4c7164309d810de660fcea3
Content hash
84a2ffa55206e037450bc323e03b121f3742950c95f5838f9b9dec8e5c6d5978
Tree hash
c74e32d0ae64fdfd5b0444b314242b6e751d77dabb9ec88a85566246eaba4a77
Skill path
skills/obra/remembering-conversations
Audit payload hash
763b7a9d629cb5ac12e38edb75b53a46

Analysis metadata

Audit model: codex

Analysis state: Complete

Scope is limited to the recorded files, lines, methods, and evidence. No runtime or sandbox execution is claimed.

Verify and export

JSON report CITATION.cff SARIF 2.1.0 Signed installation manifest Source

The manifest and lockfile bind install artifacts to cryptographic hashes. This integrity claim is separate from the security assessment.

Audit attestation: not_attestable