Skip to main content
Glama

AgentPlaybooks

by ai.agentplaybooks

Write memory

write_memory
DestructiveIdempotent

Create or overwrite a memory entry by key. There is no separate update_memory; a second write to the same key replaces the previous value and cannot be undone. Use tier='working' for active tasks, 'contextual' for background context, 'longterm' for completed work. Set memory_type='hierarchical' and parent_key to build task graphs. Requires memory:write or full permission. Use delete_memory to remove a key, archive_memories to move it to longterm without deleting, and read_memory to fetch without replacing. Pass playbook_id as the UUID or GUID of the playbook this call should target.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
keyYesMemory key
tagsNoTags for categorization
tierNoMemory tier (default: contextual)
valueYesValue to store
statusNoTask status (for hierarchical task tracking)
summaryNoCompact summary for context views
metadataNoGraph metadata: dependencies, thread assignment, progress, etc.
priorityNoPriority 1-100 (default: 50)
parent_keyNoParent memory key for hierarchical organization
descriptionNoHuman-readable description
memory_typeNoflat (default) or hierarchical for task graphs
playbook_idYesUUID or GUID of the target playbook

TDQS

A4.9/5.0
Behavior5/5

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

The description clearly discloses the consequential overwrite behavior: 'a second write to the same key replaces the previous value and cannot be undone.' It also adds permission requirements and confirms no update-only variant exists, going beyond the annotations without contradicting them.

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?

Every sentence delivers distinct, actionable information: core behavior, overwrite consequence, tier guidance, hierarchical usage, permission, and sibling alternatives. It is compact and well structured with no repetition of schema text.

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

Completeness5/5

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

For a write operation with 12 parameters, the description provides core behavior, irreversibility, permission requirements, lifecycle usage, task-graph construction, target playbook identification, and alternatives. This is sufficient context for an agent to select and invoke the tool correctly even without an output schema.

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

Parameters4/5

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

Schema coverage is 100%, so the description does not need to repeat each field. It adds valuable semantic context for tier values and the hierarchical memory_type plus parent_key combination, and it explains the intended use of playbook_id. This exceeds the baseline, though optional parameters like tags/metadata remain only schema-defined.

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 starts with a specific verb+resource: 'Create or overwrite a memory entry by key.' It explicitly disambiguates from a separate update_memory and names alternative memory operations, making the tool's role clear.

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

Usage Guidelines5/5

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

It gives concrete when-to-use guidance for tiers, hierarchical graph building, and permission requirements. It explicitly directs the agent toward delete_memory, archive_memories, and read_memory for different needs, which distinguishes it from siblings.

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

All AgentPlaybooks 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

A4.5/5.0
Disambiguation5/5

Each tool targets a distinct entity and action (e.g., delete_memory vs delete_skill vs delete_run), and even similar operations like read_memory vs search_memory vs get_memory_context have clearly differentiated purposes. The descriptions are detailed and explicitly cross-reference other tools to avoid confusion.

Naming Consistency5/5

All tool names follow a consistent verb_noun pattern (create_*, list_*, update_*, delete_*, read_*, etc.), with plurals used uniformly for list operations (list_playbooks, list_runs, list_secrets). No mixed conventions or ambiguous verbs; the naming is highly predictable and systematic.

Tool Count4/5

With 48 tools, the server covers a broad but coherent set of domains (playbooks, personas, skills, memory, canvas, runs, secrets, MCP servers, and discovery). While this exceeds the typical 3-15 range, each tool serves a distinct and necessary function within the comprehensive playbook management scope, so the count feels justified rather than bloated.

Completeness5/5

The tool surface provides complete CRUD and lifecycle coverage for every entity type: playbooks, personas, skills (including versioning and rollback), memory (including hierarchical tasks and tiering), canvas (with locking and patching), runs, secrets (including rotation and usage), and MCP servers. Additionally, find_tools covers discovery for federated tools, leaving no apparent dead ends.

Resources