Skip to main content
Glama

record_decision

Record an architectural decision with context, alternatives, and tags. Optionally lock it with do_not_revert to ensure future AI sessions respect it or get alerted on conflicting edits.

Instructions

Record one architectural decision. Set do_not_revert=true to lock it across sessions and IDEs. Returns {decision_id, session_id}. To change it later use supersede_decision (preserves the audit trail) or set_decision_flag (toggle do_not_revert / tags).

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
tagsNoOptional list of tag strings (e.g. ["security", "auth"]). Surfaces in list_decisions / list_tags filters.
forceNoIf true, skip the implicit `check_conflict` duplicate/conflict warning step. Use when you've already reviewed a conflict and want to record anyway.
symbolNoOptional function/class name within file_path to scope the decision to (e.g. "login"). With do_not_revert, the lock then blocks only edits INSIDE that symbol; edits elsewhere in the file warn instead. Requires file_path.
contextNoFree prose: why this won, what it depended on, what evidence backed it. Surfaced verbatim when a locked decision blocks an edit — this is what the next agent reads instead of guessing.
decisionYesThe decision itself (1 sentence is fine)
file_pathNoOptional file/path the decision pertains to
session_idNoOptional session id to attach to (auto-generated if omitted)
do_not_revertNoIf true, mark the decision as protected — future sessions will see do_not_revert=true and must NOT propose changes that conflict without surfacing this decision to the user first. Default false.
would_re_examine_ifNoThe condition that should trigger reconsidering this (e.g. "if the payload exceeds 1 MB" or "if we add a second write path"). Especially valuable with do_not_revert — it turns a one-way ratchet into a lock with a stated release condition.
alternatives_consideredNoThe strongest options you REJECTED, one per entry (e.g. ["polling — simpler but 3s worst-case latency", "webhooks — needs a public endpoint"]). Surfaces the losers so a future session can weigh whether to revisit instead of re-deriving them.
Behavior4/5

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

Beyond the annotations (readOnlyHint=false, destructiveHint=false), the description discloses that do_not_revert locks the decision across sessions and IDEs, and it specifies the return shape {decision_id, session_id}. It also hints at the audit trail preservation via supersede_decision. No contradiction with annotations, and useful behavioral context is added.

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?

The description is three sentences, front-loaded with the core purpose, and every sentence earns its place by covering action, locking behavior, return value, and alternatives. There is no redundant phrasing or unnecessary detail.

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?

Given the high schema coverage and that the tool has no output schema, the description compensates by specifying return values and the cross-session lock effect. It does not mention the implicit check_conflict behavior or the force parameter, but those are covered in the schema. The description is complete enough for a well-informed agent.

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?

The input schema provides 100% parameter coverage with detailed descriptions for all 10 parameters, so the description need not repeat them. It does add extra context for do_not_revert ('lock it across sessions and IDEs'), reinforcing the schema, but does not substantially expand meaning for other parameters. Baseline 3 is appropriate.

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 states 'Record one architectural decision' with a specific verb and resource, clearly distinguishing the create action from siblings. It also names supersede_decision and set_decision_flag as alternatives for later changes, explicitly differentiating this tool from those.

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 indicates when to use the tool ('Record one architectural decision') and provides explicit alternatives for later modifications ('To change it later use supersede_decision ... or set_decision_flag'), giving clear context. It does not mention exclusions or conditions when this tool should be avoided, but the guidance is sufficient for basic selection.

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

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/sachinshelke/codevira'

If you have feedback or need assistance with the MCP directory API, please join our Discord server