Skip to main content
Glama

hold_postmortem

Destructive

Hold a private postmortem meeting for an agent run or draft, then conclude it by applying accepted fix, rule, and memory artifacts.

Instructions

Hold the postmortem meeting for a draft, then conclude it. Refused while any finding's fix is pending or a task with a finding is still running (the error names what); a draft with no findings closes with no meeting. action hold (default): every agent named in a finding, every agent that did the original work in the run (or the participants you pass), and you sit at the table. opening is your account of what happened, including your own part: turn 0. The format is waffle-house when exactly one agent is responsible for every finding (it defends, the others challenge, your opening is the charge), otherwise collaborate for rounds (default 2, 1-4); format forces either. The meeting is private and takes no outside input: the human has no seat, nothing they type reaches it, and its context never carries their chat, confirmation or original request. It ends with an artifact block, one line each: Artifact: | | | . Kinds: fix (follow-up), rule and contract rule (project lessons, read before the next plan and contract), persona addendum (appended to an agent's persona file under ## Postmortem addenda, with a backup), agent rule (shown in that agent's task card), memory (the agent's memory file; memory_target: "claude" also writes the repository's .claude/agent-memory). Dissent score artifacts carry a dissent id, finding ids, an outcome (vindicated/refuted/untested) and supporting evidence; the meeting context and docs/modes/postmortem.md give the exact artifact syntax. The reviewed run’s dissents are included; scores remain advisory. Accepting one requires orchestrator (the contract’s orchestrator identity) and project opt-in; it uses the same scoring path as score_dissent. The result has the transcript, the proposed artifacts and any lines that could not be used. action conclude: pm, your closing statement, and accept[] (artifact ids) / reject[{id, why}]. Only accepted artifacts take effect, each audited. The result has a summary and the record path to relay to the human, who can read every turn (postmortem read).

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "hold_postmortem"
  }
}

Input Schema

NameRequiredDescriptionDefault
pmNoconclude: the postmortem (pm-…) hold returned.
workNoA short label for at-the-table work (for example "checkout redesign") when there is no run. Give this or execution_id.
modelNoModel id. Defaults: claude-sonnet-5-5 (anthropic), gpt-5.6-luna (openai), zai-org/GLM-5.3 (together), qwen/qwen3-235b-a22b-instruct-2507 (replicate), qwen3:14b (ollama), default (openai_compatible). Model aliases: "opus" (claude-opus-5-5), "sonnet" (claude-sonnet-5-5), "haiku" (claude-haiku-5-5). Replicate accepts owner/name, owner/name:version, or a https://replicate.com/owner/name URL.
acceptNoconclude: ids of the artifacts to apply (a1, a2, …).
actionNohold (default) or conclude.
formatNohold: force a format (default: waffle-house for a single responsible agent, else collaborate).
rejectNoconclude: artifacts you turn down, each with why.
roundsNohold: rounds (default 2).
closingNoconclude: your closing statement, the last turn of the record.
openingNohold: your account of what happened, including your own part. Do not paste the human's words.
projectNoThe project: a directory inside its git repository (resolved with git rev-parse --show-toplevel; refused outside a git work tree or with submodules), or a project key returned earlier.
base_urlNoOnly with provider "openai_compatible": base URL of an OpenAI-compatible /v1 endpoint (e.g. http://localhost:1234/v1). Overrides OPENAI_COMPATIBLE_BASE_URL. OPENAI_COMPATIBLE_API_KEY, if set, is sent only to the OPENAI_COMPATIBLE_BASE_URL origin or one listed in ROUNDTABLE_OPENAI_COMPATIBLE_TRUSTED_ORIGINS; any other base URL gets no Authorization header and must be https (http only to loopback, or to private addresses with ROUNDTABLE_ALLOW_PRIVATE_BASE_URLS=1), never a cloud metadata or (without that setting) private address.
providerNoLLM provider. Default: anthropic. Options: anthropic, openai, together, replicate, ollama (local, when reachable), openai_compatible (set OPENAI_COMPATIBLE_BASE_URL or pass base_url).
execution_idNoThe heads-down run (exe-…) the draft belongs to. Give this or work.
orchestratorNoconclude: required when accepting dissent score artifacts; the reviewed contract’s orchestrator identity, as for score_dissent.
participantsNohold: agents to seat in addition to those named in findings (for at-the-table work: everyone who did it).
memory_targetNohold: with "claude", accepted memory artifacts are also appended to <repo>/.claude/agent-memory/<agent>/MEMORY.md (refused through symlinks).

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv2.0.0

TDQS

A4.5/5.0
Behavior5/5

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

Beyond the annotations, the description discloses refusal conditions, meeting privacy ('takes no outside input', the human has no seat), artifact side effects such as appending to persona files with a backup and writing agent memory files, and that only accepted artifacts take effect and are audited. This is unusually rich behavioral disclosure for a destructive, open-world tool.

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 description is front-loaded with purpose and refusal rules, and its length is partly justified by a complex two-action workflow with many artifact kinds. However, the middle section is an extremely dense block of clauses, and the artifact-kind enumeration is hard to scan. It could be restructured into clearer hold/conclude sections without losing information.

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 17-parameter tool with no output schema and destructive annotations, the description covers both actions, refusal conditions, privacy constraints, artifact kinds列, acceptance effects, auditing, orchestrator requirements, and what each action returns. An agent has enough information to call it correctly and understand the consequences.

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 schema already documents all 17 parameters and the baseline is 3. The description adds meaningful context for several parameters: 'opening' is framed as turn 0 of the record, 'format' is tied to responsibility structure, 'rounds' has a shown default, and 'accept'/'reject' are tied to artifact ids and audit behavior. It does not add much beyond the schema for provider/model/base_url, which is acceptable because those are self-documenting.

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 opening sentence gives a specific verb and resource: 'Hold the postmortem meeting for a draft, then conclude it.' The description also distinguishes the two action modes and implicitly separates the write tool from the read-only sibling 'postmortem read' and from 'score_dissent'.

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?

It clearly states preconditions for holding a meeting: refusal while any finding's fix is pending or a task with a finding is still running, and the no-findings case. It also names the alternative scoring path for dissent artifacts and mentions 'postmortem read' for the human. It does not explicitly compare against other meeting siblings such as end_meeting or end_local_meeting.

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