session-sitter

configuring-session-sitter

Set up Session Sitter for a user, or change an existing configuration. Use when someone has just installed the extension, asks how to turn on supervision, Telegram cards or remote control, wants auto-approve rules or workspace colours, asks why a setting is not taking effect, or wants their configuration reviewed. Interviews the user, writes the settings, and validates the result with a script rather than from memory.

checking-a-call-against-policy

Find out whether Session Sitter will allow a tool call, and which written clause decides it, BEFORE running it. Use when about to run something plausibly governed — a push, a deploy, a migration, anything touching secrets or infrastructure — or when the user asks what the policy allows, or why a call was denied.

reading-the-audit-trail

Read and interpret the Session Sitter audit trail — what agents were allowed to do, which clause was applied, who decided, and how long it took. Use when the user asks what happened overnight, why a call was denied or rewritten, which rules are firing, or wants to export the decision log.

writing-practices

Write or edit a Session Sitter practices file — the markdown that decides permission prompts. Use when the user wants to add a rule, block a command, allow something without approval, set up a practices file, or asks why a clause did not fire.

kb-sitter

Use when a supervision runtime needs to load BDI knowledge (three bottom-line.md files) for one running coding-agent session. Given a (user, project, team) triple and a knowledge repo, this skill fetches the three tier files and returns them in tier order (team → project → user) for merging. Do not invoke for general repo browsing, plain file reads, or for classifying observed agent actions.