claude-task-runnerRun Claude Code tasks in headless mode with `claude -p` through the local `cc-task-runner.sh` wrapper, including model switching, JSON output capture, struct...
Install via ClawdBot CLI:
clawdbot install zwd0313/claude-task-runnerGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Calls external URL not in known-safe list
https://open.bigmodel.cn/api/anthropic`Audited Apr 18, 2026 · audit v1.0
Generated Oct 7, 2026
A DevOps team triggers cc-task-runner.sh in CI to run Claude Code headless reviews on every pull request using GLM-5.1. Results are enforced against a JSON schema so graders can automatically reject PRs that fail structured output checks. Artifact flags verify that review reports are written to the expected path.
Financial analysts use run-file with minimax-m2.7 to process large prompt files that generate weekly market summaries in headless mode. The --expect-file and --expect-glob flags confirm each report artifact was created before downstream dashboards ingest them.
A platform team uses the multi-file validation pattern to have Claude Code generate service boilerplate, config files, and tests in one run. --expect-contains binds each generated file to critical content like package names or API endpoints, catching partial or malformed generations automatically.
Legal operations teams run scripted Claude Code tasks that produce structured audit evidence files for regulatory reviews. The durable task state under ~/.openclaw/state/cc-tasks/ plus log inspection provides a defensible record of each non-interactive execution.
Marketing teams generate landing page copy and asset manifests in multiple languages by running cc-task-runner.sh with schema-enforced output. Glob checks ensure every locale directory contains generated files before publishing to the CMS.
Sell a managed wrapper around headless Claude Code execution that adds retries, fallback models, artifact validation, and audit logs for enterprise engineering orgs. It plugs into existing CI systems and replaces fragile shell scripts with a governed runner.
Offer a hosted service where regulated industries run Claude Code tasks with enforced JSON schemas and immutable artifact verification trails. Customers pay for guaranteed auditability of every automated action performed by the AI agent.
Provide implementation services that wire cc-task-runner.sh into client CI/CD pipelines, define JSON schemas, and tune model selection between GLM-5.1 and minimax-m2.7. Deliverables include runner configurations, validation rule libraries, and training for internal teams.
💬 Integration Tip
Wrap cc-task-runner.sh calls in your CI or orchestration layer and always pair a JSON schema with at least one artifact check so failures are caught by real outputs, not model text alone.
Scored Oct 7, 2026
Real-time search engine supporting web search, vertical domain search, parallel batch search, and URL content extraction.
Manage Feishu (Lark) calendars by listing, searching, checking schedules, syncing events, and marking tasks with automated date extraction.
cad reference tool
Search, install, and create OpenClaw skills using intelligent matching across built-in, local, and GitHub skill repositories.
Use when building CLI tools, implementing argument parsing, or adding interactive prompts. Invoke for CLI design, argument parsing, interactive prompts, progress indicators, shell completions.
Publish local skills to ClawHub with customizable version, name, path, and optional changelog using command line or Python API.