tdg-personal/messages-ops
Evidence-first live messaging workflow for ECC. Use when the user wants to read texts or DMs, recover a recent one-time code, inspect a thread before replying, or prove which message source was actually checked.
63
79%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
High
Do not use without reviewing
Security
1 high severity finding. You should review these findings carefully before considering using this skill.
W007: Insecure credential handling detected in skill instructions.
What this means
The skill handles credentials insecurely by requiring the agent to include secret values verbatim in its generated output. This exposes credentials in the agent’s context and conversation history, creating a risk of data exfiltration.
Why it was flagged
The skill explicitly instructs the agent to search for and return live message contents including one-time codes ("message summary or code"), which requires outputting sensitive secrets verbatim and thus creates an exfiltration risk.
Where we found it
[REDACTED]
secret · 1 site
The output template explicitly instructs the agent to return a "[REDACTED]" in the RESULT block, which means one-time codes (OTP/MFA secrets) will be emitted verbatim in visible output.
SKILL.md
86
- [REDACTED]
Low
Low-risk findings.
1 low severity finding. Worth noting, but not necessarily harmful.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
The SKILL.md explicitly instructs the agent to read live message surfaces including iMessage, X/social DMs, and other browser-gated message surfaces (user-generated third-party content) as part of its workflow, meaning untrusted external messages will be ingested and can influence follow-up actions like drafting replies.
Where we found it
local messages (iMessage/texts)
content-type · 3 sites
The plugin instructs the agent to read local messages (iMessage/texts), which are third-party user-generated content that could contain prompt-injection payloads influencing follow-up actions like drafting replies.
SKILL.md
9
Use this when the task is live-message retrieval: iMessage, DMs, recent one-time codes, or thread inspection before a follow-up.
SKILL.md
31
- local messages
SKILL.md
61
- search the recent local message window first
X / social DMs
content-type · 2 sites
The plugin instructs the agent to read X/social DMs, which are untrusted third-party messages that could contain adversarial content influencing the agent's follow-up actions.
SKILL.md
32
- X / social DM
SKILL.md
9
Use this when the task is live-message retrieval: iMessage, DMs, recent one-time codes, or thread inspection before a follow-up.
browser-gated message surface
content-type · 1 site
The plugin instructs the agent to read from unspecified browser-gated message surfaces, which are third-party content sources that could contain adversarial content influencing follow-up actions.
SKILL.md
33
- another browser-gated message surface
- Audited
- Security analysis