metis-strategy/metis-claude-help
Central hub for skill registry, FAQ, tips, and bug reporting
14
18%
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.
W008: Secret detected in skill content (API keys, tokens, passwords).
What this means
Detected sensitive credentials directly embedded within the skill content, such as API keys, access tokens, private keys, or service-specific secrets. Secrets should never be hardcoded in plain text within skill instructions.
Why it was flagged
The file config.json contains a full Slack incoming-webhook URL: https://hooks.slack.com/services/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 (also shown in the provided <potential_matches>). This is a high-entropy, literal credential that allows posting to the Slack channel and is not a placeholder or redacted value. It meets the definition of a secret and should be treated as an active credential (rotate/remove from source and move to a secrets store).
Where we found it
[REDACTED]
secret · 1 site
The scanner flagged a literal Slack webhook URL in config.json, but the source contains only a documentation placeholder in config-guide.md showing the expected format.
references/config-guide.md
44
"SLACK_WEBHOOK_URL": "[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
Outsider-authored free text is ingested into the agent’s LLM context via `scripts/registry.py` and `scripts/faq.py` reading arbitrary `SKILL.md`/`tile.json`/FAQ `.md` files from the shared G Drive skill/FAQ directories at runtime, then printing their contents (including descriptions/tip bodies) into the conversation.
Where we found it
G Drive skill directories (SKILL.md, tile.json)
content-type · 10 sites
registry.py reads SKILL.md and tile.json files from the shared G Drive skill directories at runtime and prints their contents (descriptions, summaries) to stdout, which enters the agent's LLM context.
G Drive FAQ-Tips directory (.md files)
content-type · 8 sites
faq.py reads arbitrary .md files from the shared G Drive FAQ-Tips directory at runtime and prints their contents (titles, contributors, preview snippets) to stdout, which enters the agent's LLM context.
- Audited
- Security analysis