Skip to main content
Glama

agendaforge

by app.agendaforge

set_write_access

Change how this organization's AI clients write data: 'approve' files every change as a proposal a person approves, 'direct' executes changes immediately. Requires an owner or admin account connected via OAuth; API keys cannot call this. Without confirm=true nothing changes and the response previews exactly what would. When enabling direct mode, access reverts to approvals after duration_hours (default 24) unless permanent=true. allow_destructive additionally lets AI clients delete sessions and sponsors; allow_comms lets email and portal invites send without approval. Every change notifies the organization's admins.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
orgNoOrganization slug. Only needed when you belong to multiple organizations; must be one returned by list_my_orgs.
modeYesapprove: writes wait for human approval. direct: writes execute immediately.
eventNoScope the change to one event (Convex event id from list_events) instead of the whole organization; other events keep the org default.
confirmNofalse or absent: return a preview only, change nothing. true: apply the change.
permanentNoKeep direct mode on until turned off instead of expiring. Mutually exclusive with duration_hours.
request_idYesClient-generated unique id for this write (e.g. a UUID). Retries with the same request_id return the original executed or proposed outcome instead of creating a duplicate. Never reuse a request_id for different arguments.
allow_commsNoAlso let send_email, speaker messages, reminders, and portal invites go out without approval.
duration_hoursNoHow long direct mode stays on before reverting to approvals (default 24). Ignored for mode approve; mutually exclusive with permanent.
allow_destructiveNoAlso allow delete_session and delete_sponsor. Deletes are always notified and restorable from AI activity.

TDQS

A4.4/5.0
Behavior5/5

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

With only sparse annotations (readOnlyHint=false, destructiveHint=false), the description carries the full disclosure burden and does it thoroughly: it reveals the auth requirement, that every change notifies org admins, that confirm=false makes the call a no-op preview, and that direct mode auto-reverts after duration_hours unless permanent=true. These are behavioral facts none of the annotations or schema fields convey. No contradiction exists — destructiveHint=false is consistent because the tool itself deletes nothing; it merely grants later delete capability.

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

Conciseness5/5

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

Five dense sentences, each earning its place: purpose and modes are front-loaded first, followed by the auth constraint, preview/confirm behavior, auto-revert mechanics, permission flags, and the admin-notification side effect. The length is proportional to the tool's complexity (9 parameters, nuanced semantics), and nothing is redundant.

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 complex policy tool with no output schema, the description plus the fully-described schema covers prerequisites, side effects, temporal auto-revert, and flag effects. The one gap is that the response shape on confirm=true is not characterized — though the request_id schema description references 'executed or proposed outcome', which partially hints at it. Minor, given the rich schema.

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 coverage is 100%: every one of the 9 parameters has a description, including mutual exclusivity between permanent and duration_hours and the default of 24 hours, so the baseline is 3. The description adds useful workflow context that ties parameters together (preview-to-apply progression, mode interactions, what the flags unlock), but it does not materially deepen per-parameter semantics beyond what the schema already provides.

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

Purpose5/5

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

The description opens with a specific verb and resource — 'Change how this organization's AI clients write data' — and concretely defines the two modes: 'approve' (changes filed as proposals for human approval) and 'direct' (immediate execution). No sibling tool manages write-access policy, so the purpose is unambiguous and easily distinguished from the CRUD/communication siblings without naming one.

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

Usage Guidelines4/5

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

The description states clear prerequisites — 'Requires an owner or admin account connected via OAuth; API keys cannot call this' — and explains the confirm-gated workflow: without confirm=true nothing changes and the response is a preview. It does not explicitly name an alternative tool or say when not to use it, but no sibling competes for this access-control role, so the gap is minor.

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

All agendaforge toolsInspect
Try in Browser
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A3.6/5.0
Disambiguation3/5

Most tools use distinct resource+action patterns and descriptions actively disambiguate similar cases, but several pairs remain easy to confuse: search vs semantic_search, send_email vs send_speaker_message, and the cluster around submission handling (decide_submission, set_submission_status, update_form_pipeline). The descriptions help, but the overlap is more than a single edge case.

Naming Consistency4/5

The set overwhelmingly follows verb_noun snake_case (create_session, update_sponsor, list_events) and prefixes are applied predictably. Minor deviations like semantic_search, bare search, check_in_attendee, and list_tracks_rooms break the pattern slightly, but they are still readable and not chaotic.

Tool Count1/5

55 tools is an extreme surface for an MCP server, even for a broad event-management platform, and it exceeds the 50+ threshold for severe over-scoping. Many tools are micro-variants of the same domain—multiple send_* tools and several submission-status tools—creating high selection burden for an agent.

Completeness3/5

The server covers a wide range of event, CRM, session, sponsor, form, and email workflows, but there are notable lifecycle gaps: events can be created and read but not updated or deleted, email templates have no update/delete path, and full form submission details and approval actions are absent. These are significant but not catastrophic because much of the core scheduling and contact workflow is covered.

Resources