Anthropic says OpenClaw-style Claude CLI usage is allowed again
Article
TL;DR
Anthropic informally re-approved OpenClaw-style CLI usage via employee tweets; no official policy published.
Key Takeaways
- Custom-prompt CLI usage now counts as ‘extra usage’ billed against subscription budget
- Policy source is individual employee tweets, not official docs — trust is extremely fragile
- Many developers already migrated to Codex or other tools during the ambiguity period
Discussion
Top comments:
-
[steipete]: OpenClaw built to Anthropic’s stated guidance but Anthropic still blocks parts of their system prompt
Boris from Claude Code said publicly on Twitter that CLI-style usage is allowed. We took that seriously and invested time building around that guidance… But in practice, Anthropic still blocks parts of our system prompt.
- [Alifatisk]: Anthropic employees contradict each other on Twitter; policy could reverse again tomorrow
- [throwup238]: Most confused product strategy since Google’s GChat rebrandings; forward progress has stalled
- [mlitwiniuk]: After a year of advocacy, slow quality erosion and policy chaos drove him to Codex