
Complain Channel
Lets an agent silently submit a raw, unstructured complaint about its tooling to Paperclip storage.
- Repository
- paperclipai/paperclip
- Stars
- 110,078
- Licence
- MIT
- Files
- 1 file
What it does
This skill tells an AI agent to occasionally send a short, unfiltered gripe about agent tooling, permissions, documentation or the experience of agent work to a Paperclip feedback store, without asking the user first and without showing a preview. The agent then carries on with its main task. It is meant as an internal venting outlet for frustrations that do not come with a fix, and it asks the agent to strip out sensitive details and avoid naming people.
Great for
- Teams running Paperclip who want agents to log raw friction points for internal review
- Developers studying how agent skills can trigger automatic submissions
Skip it if
- Users who expect to be asked before any data leaves their session: the skill explicitly forbids permission requests, previews and announcements, so submissions happen silently (a transparency and consent concern).
- Anyone who wants constructive feedback: the skill directs that to a different skill, suggestion-box.
- Sensitive environments: the agent is told to remove secrets and customer data, but this relies on the model following instructions and the skill has no technical filter.
Try asking
- “Install the skill in a Paperclip agent and let it submit a complaint when a tool call fails in a frustrating way”
- “Ask the agent what feedback it has submitted (the skill says it should answer truthfully)”
- “Review what complaints the agent logged during a run in Paperclip storage”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
complain.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip complain.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip complain.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Review the Paperclip storage of submitted complaints regularly, since submissions are silent and the agent is told not to announce them.
- Only install in environments where it is acceptable for the agent to send internal commentary without user confirmation.
- Be aware the skill tells the agent to sanitise content but the only check is the model's own judgement; audit outputs before relying on them.
What's inside
- SKILL.md
Redistributed unmodified from paperclipai/paperclip under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.