Skip to main content
Glama

skein_watchers

The caller's own notification subscriptions, at three scopes: one issue, a whole project, or a saved filter (custom watcher). Automatic subscriptions: the assignee of an issue (except one filled in at create by the "assign agent-created issues to me" setting); its creator (the human who owns the API key) only for issues created by hand or as a TOP-LEVEL agent_directed issue — an issue your agent files under a parent, or with source=agent_autonomous, does not subscribe you. Unsubscribing from an issue you watch records an opt-out: the auto-watch paths (creator, assignee, reopen) will not re-subscribe you to that issue until you watch it again with action="watch", and the issue's card leaves your Watching column. Per-user cap: 10 filter watchers. Also stops one of your task monitors (action="monitor_stop", by the monitorId skein_get_monitor_script returned). Read-only actions: list, filter_list.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "action": "<action>"
    },
    "name": "skein_watchers"
  }
}

Input Schema

NameRequiredDescriptionDefault
nameNoFilter-watcher name, unique per user, 100 chars max. Required for filter_create.
actionYesWhat to do: - "watch" → subscribe to one issue (reason='manual'; removes your opt-out from an earlier unwatch). - "unwatch" → unsubscribe from one issue: the subscription is deleted, an opt-out stops automatic re-subscription, the issue's Watching card is cleared. Not subscribed → only the card is removed, no opt-out. - "project_watch" → subscribe to a whole project's notifiable events. - "project_unwatch" → unsubscribe from a project. - "filter_create" → create a saved-filter watcher from `name` + `filter`. - "filter_delete" → remove one filter watcher by `filterId`. - "filter_list" → list the caller's filter watchers. - "list" → the caller's full subscription state: issues, projects and filters at once. - "monitor_stop" → stop one of your task monitors by `monitorId`: its token is revoked, its runner ends with token_expired on its next poll, other monitors (same issue included) keep running. Works from any of your API keys; stopping an already stopped monitor succeeds; another user's or an unknown id → monitor_not_found.
filterNoFilter-watcher matching rules. Required for filter_create. All fields optional, an omitted field means "no constraint": event_types[] | projects[] | types[] | labels[] | components[] | priority_max.
issueIdNoFull hierarchical issue id, e.g. "demo-42" or "demo-42.1". Required for watch and unwatch.
projectNoProject handle: alias, name, or UUID. Required for project_watch and project_unwatch.
subtreeNoOnly read by watch. When true, subscribe to this issue AND its subtree by the hierarchy: every issue that has it as a parent, grandparent, … at the moment of each event (parentIdCurrent chain — not an id prefix); future subtasks are covered automatically. subtree dominates: it upgrades an existing plain watch and a later subtree=false watch never downgrades it. Default: false (this issue only).
filterIdNoFilter-watcher UUID, as returned by filter_create or filter_list. Required for filter_delete.
monitorIdNoMonitor id (UUID) as returned by skein_get_monitor_script. Required for monitor_stop.
eventTypesNoOnly read by project_watch: whitelist of event types to subscribe to (e.g. ["issue.created", "comment.created"]). Each value must be a notifiable event type. Omit to leave the existing whitelist alone (or receive every notifiable type on a first subscribe); pass an empty array to clear it back to "all".

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A3.8/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The annotations only supply a bare safety profile (readOnlyHint=false, destructiveHint=false, idempotentHint=false); the description carries the real behavioral load. It discloses non-obvious rules an agent could not otherwise predict: which auto-subscriptions exist and their carve-outs ("except one filled in at create..., an issue your agent files under a parent, or with source=agent_autonomous, does not subscribe you"), that unwatch records a persistent opt-out, the 10-filter-watcher cap, and monitor_stop semantics including token revocation, cross-key operation, and the monitor_not_found error.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The first sentence is front-loaded and the read-only/write split is called out, which is good structure. But the body is a dense run-on with deep parentheticals (the auto-assign and TOP-LEVEL agent_directed qualifiers) and a monitor_stop tangent bolted on as "Also stops one of your task monitors," which belongs with the action list rather than the subscription narrative.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 9-parameter, multi-action tool with a nested filter object and no output schema, the definition covers scopes, auto-subscription rules, opt-out persistence, caps, and an error case, which is enough to invoke it correctly. The remaining gap is return-shape information: the description never says what a list or filter_list response contains beyond the schema's one-line gloss, and no output schema exists to compensate.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100% and the schema already documents every action, id formats, subtree dominance, and eventTypes whitelist behavior in depth. The description adds essentially one parameter-relevant fact beyond the schema, the per-user cap of 10 filter watchers, and otherwise restates schema content. With the schema doing the heavy lifting, the baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The opening sentence names the resource and scope precisely: "The caller's own notification subscriptions, at three scopes: one issue, a whole project, or a saved filter." That is far more specific than the tool name alone, and the phrase "the caller's own" establishes the ownership boundary. It stops short of naming a sibling such as skein_notifications, so an agent cannot tell from the description alone how this differs from the broader notification tool.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives detailed conditions for individual behaviors, e.g. that unsubscribing "records an opt-out" and the auto-watch paths will not re-subscribe, which helps an agent understand consequences. However, it never states when to choose this tool over an alternative sibling tool, and there are no explicit when-not-to-use exclusions. Usage is therefore implied by the enumerated scopes rather than guided.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.