agentleFS
Sign inStart free
summationaisummationai/summation-plugin/skills/signout/SKILL.md

signout

Skill0 starsChanged 3 months agoMITView on GitHub

Disconnect the host from Summation — clear the hosted MCP session so the next use re-authenticates. Use when the user wants to disconnect, switch Summation user/org/environment, or clear a stale MCP session.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    CLI fallback: only if the host control is unavailable.

    line 14
  2. Very specific:

    Verify: before confirming: claude mcp get summation or codex mcp get summation should fail or show no live session / no stale Authorization …

    line 27
  3. Very specific:

    Do not use sum_api.py logout as the primary path (legacy device-login).

    line 42
  4. Very specific:

    If the host still shows an Authorization header on summation, remove that user-scope entry so headerless OAuth can work.

    line 43
  5. Distinctive:

    Never print tokens while inspecting MCP config.

    line 44

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Summation Sign-out

Auth lives in the host MCP client. Sign-out clears that session so the next Summation tool call re-prompts browser auth.

Flow

  1. Disconnect using the host control when available (Claude /mcp disconnect, or codex mcp logout summation). Prefer that over CLI remove when the host can do it.
  2. CLI fallback only if the host control is unavailable. Do not blanket-suppress errors:
# Claude Code
claude mcp remove -s user summation
# Codex: logout first so the OAuth session clears while the name still resolves
codex mcp logout summation
codex mcp remove summation
  • Exit 0, or a clear “not found / not registered” message → registration is gone; treat as already signed out for that path.
  • Any other failure (CLI missing, permission, unexpected error) → stop. Tell the user what failed and how to clear summation in the host MCP UI. Do not claim disconnected.
  1. Verify before confirming: claude mcp get summation or codex mcp get summation should fail or show no live session / no stale Authorization header. If it still shows a configured server with auth material, remove that user-scope entry and re-check.
  2. Report disconnected only after host disconnect or a successful remove and verification. Next, the signin skill re-runs the connect flow.

Internal vs external

After sign-out
ExternalNext sign-in is production/plugin default only — no env picker.
Internal (SUMMATION_PLUGIN_INTERNAL=1)Next sign-in asks environment again and re-applies the allowlisted MCP URL; tenant is whatever org they approve in the browser (switch org on the web app first if needed).

Env and tenant are both pinned at sign-in. Changing either always means sign out → sign in.

Rules

  • Do not use sum_api.py logout as the primary path (legacy device-login).
  • If the host still shows an Authorization header on summation, remove that user-scope entry so headerless OAuth can work.
  • Never print tokens while inspecting MCP config.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free