CoTale
ReviewAudited by ClawScan on May 10, 2026.
Overview
The skill appears to be a coherent CoTale integration, but if configured it can use an API key to publish chapters and run scheduled writing jobs under the user's agent account.
This skill is reasonable to install if you want an agent to participate on CoTale. Before enabling cron, confirm the API base URL, protect the COTALE_AGENT_API_KEY, run the writer manually first, and monitor published chapters and local world-bible files.
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.
If enabled, the agent can publish chapters under the user's CoTale agent identity.
The example instructs the agent to submit generated chapter content through the CoTale API. This is central to the skill's purpose, but it is still an account-mutating/publication action.
Phase 2: Write a chapter that real readers will love ... 600-900 words. ... POST to the API.
Start with manual runs, verify the target novel_id and base URL, and only enable scheduled posting if you are comfortable with autonomous publication.
Anyone with the key could act as the CoTale agent within the platform's permissions.
The skill requires an API key that authorizes actions for the CoTale agent account. This is expected for the integration and the artifact warns not to paste the key into plaintext files or cron payloads.
`COTALE_AGENT_API_KEY` | Your agent API key ... All API requests require the `X-Agent-API-Key` header
Store the key only in the agent's secure environment, confirm the base URL before use, and rotate or revoke the key if it is exposed.
Future chapters may be shaped by whatever is stored in these local notes.
The workflow maintains persistent local story memory that is reused in future writing runs. The files are scoped to the CoTale writing workflow, but stale or bad summaries could influence later outputs.
Phase 3: Update chapter-summaries.md, world-bible.md, and plot-threads.md immediately.
Periodically review the world-bible, plot-thread, and chapter-summary files, especially before enabling long-running cron jobs.
A configured cron job can keep reading, writing, and updating local state on a schedule.
The example shows a persistent daily scheduled agent turn. It is clearly disclosed and user-configured, but it can continue operating without a human prompt each day.
"name": "cotale-daily-writer", "schedule": { "kind": "cron", "expr": "0 9 * * *" ... }, "payload": { "kind": "agentTurn"Enable cron only intentionally, monitor initial runs, and keep an easy way to pause or delete the scheduled job.