The Agent Skills Directory

[DYNAMIC_EXECUTION]: The script scripts/gsc_diagnostics.py uses pickle.load() to deserialize local authentication tokens. This is an inherently unsafe operation because the pickle module is not secure against erroneous or maliciously constructed data. An attacker who can write a malicious payload to the token file (~/gsc_token.pkl or the path in GSC_TOKEN_PATH) can achieve arbitrary code execution when the skill is run.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the Google Search Console API, such as search queries and page URLs.
Ingestion points: The query_gsc() function in scripts/gsc_diagnostics.py fetches data from the searchanalytics.query endpoint, which includes user-controlled search queries.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when the agent processes or reports these queries.
Capability inventory: The skill uses Bash, Write, Glob, WebSearch, and Read tools, providing a wide surface for an injected command to interact with the environment.
Sanitization: The script does not sanitize or validate the content of the queries or URLs before printing them to standard output or writing them to the JSON output file.

seo-gsc-diagnostics