corpus
v1.0.1Access a user's Corpus library from OpenClaw. Use when the user asks to search saved content, fetch item details, save links into Corpus, or create reminders from Corpus content.
Installation
Please help me install the skill `corpus` from SkillHub official store.
npx skills add doninocode/corpus
Corpus Skill
Use this skill to read and write a user's Corpus data through the Corpus API.
Required environment variables
CORPUS_API_TOKEN: user token for Corpus API access.
Generate CORPUS_API_TOKEN
- Install Corpus AI for iPhone:
https://apps.apple.com/us/app/corpus-ai/id6748364607 - Open the app and sign in.
- Go to
Integrations->OpenClaw. - Create an API token and copy it (the full value is shown once).
- Set that value as
CORPUS_API_TOKENin your OpenClaw skill env.
Optional environment variables
CORPUS_API_BASE_URL: API base URL (default:https://corpusai.app).CORPUS_TIMEOUT_SECONDS: HTTP timeout in seconds (default:30).
OpenClaw config example
{
"skills": {
"entries": {
"corpus": {
"path": "/absolute/path/to/skills/corpus",
"env": {
"CORPUS_API_TOKEN": "csk_live_or_jwt_token_here",
"CORPUS_API_BASE_URL": "https://corpusai.app"
}
}
}
}
}
Commands
Run all commands through:
python3 {baseDir}/scripts/corpus_api.py <command> [options]
Available commands:
profilelist-content --limit 20 --cursor <cursor>search --query "<text>" --limit 8content --user-content-id <id>save-url --url <url> [--user-note "<note>"]create-reminder --title "<title>" --description "<desc>" --scheduled-date-utc "2026-02-18T16:00:00Z" [--user-content-id <id>]
Recommended workflow for implementation tasks
When a user asks for "find items in Corpus and implement":
- Use
searchwith a focused query. - Use
contentfor top hits to collect concrete steps. - Produce an implementation plan with explicit file changes.
- Apply code changes in the current working repository after user confirmation.
Safety rules
- Never print or log
CORPUS_API_TOKEN. - Prefer read operations before write operations.
- Before write operations (
save-url,create-reminder), confirm user intent if the instruction is ambiguous.