GLM-OCR-Handwriting
Analysis
This OCR skill is coherent and transparent, but it sends user-selected images or PDFs to ZhiPu's GLM-OCR API using your API key.
Findings (4)
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.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
ONLY use GLM-OCR API ... NEVER parse handwriting yourself ... NEVER offer alternatives ... NO fallback methods
The skill gives strong control instructions that force the GLM-OCR workflow and prohibit fallback methods. This appears intended to keep OCR behavior consistent, but it also limits the agent from suggesting local or alternative approaches.
"Error: 'requests' is required. Please install it:\n" " pip install requests\n"
The script depends on the external requests package, while the artifact set has no install spec or pinned dependency file. The dependency is common and purpose-aligned, but users may need to install it separately.
Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.
api_key = _get_env("ZHIPU_API_KEY") ... headers = {"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"}The skill uses a provider API key as a bearer token. This is expected for the ZhiPu OCR integration and is sent to the fixed provider endpoint, but users should recognize that the key authorizes API usage.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
data = path.read_bytes(); b64 = base64.b64encode(data).decode("ascii") ... requests.post(api_url, json=payload, headers=headers, timeout=timeout)For local-file OCR, the script reads the selected file, encodes it, and sends it in a JSON request to the external GLM-OCR API. This is central to the skill's purpose and is disclosed, but it may involve private document contents.