agentleFS
gokite-aigokite-ai/passport-skills/manage-agents/SKILL.md

manage-agents

Skill2 starsChanged 4 months agoMITView on GitHub

Inspect registered agents and spending sessions on the user's account. Invoke for diagnostics, debugging "agent not registered" errors, reviewing active or expired sessions, checking a separate session's budget remaining, or when the user asks what agents or sessions exist. Read-only -- does not create or modify anything. For the account allowance (set up or not, left, reset) use `kpass allowance status` instead.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

18 rules. About 0 are common, 4 are distinctive, and 14 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 14

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Prerequisite: Before using any Passport skill, check if kpass is installed by running command -v kpass.

    line 17
  2. Very specific:

    If you need to create a new session or register an agent, use the request-session skill instead.

    line 31
  3. Very specific:

    If you need to execute a payment through a session, use the x402-execute skill.

    line 32
  4. Very specific:

    If you need to list sessions from the agent's perspective (using the agent token), use session list from the request-session skill.

    line 33
  5. Distinctive:

    No agent registration or spending session is required.

    line 39
  6. Distinctive:

    Read that file before running any command.

    line 59
  7. Very specific:

    Use the authenticate-user skill.

    line 78
  8. Distinctive:

    Do NOT attempt any of the following.

    line 90
  9. Distinctive:

    Before running any command, verify:

    line 109
  10. Very specific:

    Agent ID (--agent-id): If provided, must be a string from a prior command's output.

    line 112
  11. Very specific:

    Agent type (--agent-type): If provided, must be a known agent type string (e.g., claude, cursor, codex, cline).

    line 113
  12. Very specific:

    Session ID (--session-id): If provided, must be a string from a prior command's output.

    line 114
  13. Very specific:

    Status (--status): If provided, must be active or expired.

    line 115
  14. Very specific:

    Limit (--limit): If provided, must be an integer between 1 and 100.

    line 116
  15. Very specific:

    Offset (--offset): If provided, must be a non-negative integer (0 or greater).

    line 117
  16. Very specific:

    To register an agent or create sessions: Use the request-session skill.

    line 124
  17. Very specific:

    To execute payments: Use the x402-execute skill (pays from the account allowance, or from a separate session the user approved).

    line 125
  18. Very specific:

    To check wallet balance or send tokens: Use the wallet-send skill.

    line 126

≈ “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

Manage Agents

Prerequisite: Before using any Passport skill, check if kpass is installed by running command -v kpass. If it returns nothing, invoke the passport-setup skill first to install the CLI.

List and inspect registered agents and their spending sessions from the user's perspective. These commands use the user's JWT (not an agent token) and show data across all agents registered to the user's account.

When to Use This Skill

  • The user asks "what agents are registered on my account?" or "show me my agents."
  • The user asks "what sessions do I have?" or "show me active sessions."
  • You need to debug an "agent not registered" or "no active sessions" error.
  • You want to verify an agent registration or session was created successfully.
  • The user asks for a history of agent sessions or wants to review session activity.

When NOT to Use This Skill

  • If you need to create a new session or register an agent, use the request-session skill instead. This skill is read-only.
  • If you need to execute a payment through a session, use the x402-execute skill.
  • If you need to list sessions from the agent's perspective (using the agent token), use session list from the request-session skill.

Prerequisites

The user MUST be authenticated before using this skill. If not logged in (exit code 3 with "Not logged in"), use the authenticate-user skill first.

No agent registration or spending session is required. These commands operate with the user's JWT directly.

Defaults (Do Not Ask the User Unless They Specify Otherwise)

SettingDefault valueOverride
Output format--output jsonAlways use JSON output. Never omit this flag.
FiltersOmit all filters (returns everything)Only pass filter flags when the user asks to narrow results.
LimitServer default (omit)Only pass --limit if the user requests pagination or you need to limit results.
Offset0 (omit)Only pass --offset for pagination when combined with --limit.
Base URLOmit (uses built-in default)Only pass --base-url if the user explicitly provides a custom backend URL.

Command Reference and Worked Examples

Full argument tables, JSON output examples, error envelopes for me, user agents, and user sessions — plus four end-to-end worked examples (list all agents, filter sessions by status, paginated listing, diagnose "agent not registered") — live in:

→ @references/commands.md

Read that file before running any command.


Error Handling

Exit CodeMeaningError Message PatternRecovery Action
0Successstatus: "success"Present the result to the user.
1Network errornetwork error: ...Check connectivity. Retry after a brief pause.
2Usage error--limit must be an integer between 1 and 100, --offset must be a non-negative integerFix the flag value. See validation rules in the arguments table.
3Auth errorNot logged in. Run ...Use the authenticate-user skill to log in.
4Not foundnot foundThe requested resource does not exist. Check the filter values.
5Rate limitedrate limitWait 30 seconds, then retry.
6Session policy violationN/A for manage-agentsRead-only commands do not use spending sessions. This exit code is not expected.

Specific Error Scenarios

"Not logged in." (exit code 3):

  • Use the authenticate-user skill. After logging in, retry the command.

"--limit must be an integer between 1 and 100" (exit code 2):

  • The --limit value must be a whole number from 1 to 100. Do not pass 0, negative values, or decimals.

"--offset must be a non-negative integer" (exit code 2):

  • The --offset value must be a whole number that is 0 or greater. Do not pass negative values or decimals.

Commands That DO NOT Exist

Do NOT attempt any of the following. They will fail:

  • kpass user (without a sub-command) -- must use user agents or user sessions
  • kpass user agent (singular) -- does not exist; use user agents (plural)
  • kpass user session (singular) -- does not exist; use user sessions (plural)
  • kpass user agents --status -- the --status flag only exists on user sessions, not user agents
  • kpass user agents --limit -- the --limit flag only exists on user sessions, not user agents
  • kpass user sessions --type -- the flag is --agent-type, not --type
  • kpass user sessions --id -- the flag is --session-id, not --id
  • kpass user delete-agent -- does not exist
  • kpass user revoke-session -- does not exist
  • kpass agents -- does not exist; use user agents
  • kpass sessions -- does not exist; use user sessions
  • Any command with --json -- the correct flag is --output json (two separate tokens)

Input Validation Checklist

Before running any command, verify:

  1. Authentication: The user must be logged in. Use me --output json to check.
  2. Agent ID (--agent-id): If provided, must be a string from a prior command's output. Do not fabricate values.
  3. Agent type (--agent-type): If provided, must be a known agent type string (e.g., claude, cursor, codex, cline).
  4. Session ID (--session-id): If provided, must be a string from a prior command's output. Do not fabricate values.
  5. Status (--status): If provided, must be active or expired.
  6. Limit (--limit): If provided, must be an integer between 1 and 100.
  7. Offset (--offset): If provided, must be a non-negative integer (0 or greater).

Cross-Skill References

  • Prerequisite: User must be logged in. Use the authenticate-user skill.
  • To register an agent or create sessions: Use the request-session skill. This skill is read-only and cannot create or modify agents or sessions.
  • To execute payments: Use the x402-execute skill (pays from the account allowance, or from a separate session the user approved).
  • To check wallet balance or send tokens: Use the wallet-send skill.

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.