An internal control for running one named member of an agent team or execution run. It can start, continue, correct, or dismiss that member using brief and report files.

Use it to manage individual delegated agents: give them work, send their next brief, amend instructions, or stop them.

It keeps each member operation separate from the larger plan, run record, and tracking process. This is for command workflows, not ordinary conversation.

member-run — Skill for Claude Code by restarter/lets-workflow

Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code

/plugin marketplace add restarter/lets-workflow

Claude Code

/plugin install lets

Made for: Claude Code.

Or install lets, the plugin that ships this one along with the rest of its 20 skills, 26 commands, 16 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/restarter/lets-workflow/member-run/github.svg)](https://agentmods.dev/skills/restarter/lets-workflow/member-run)

Your own site

<a href="https://agentmods.dev/skills/restarter/lets-workflow/member-run"><img src="https://agentmods.dev/badge/skills/restarter/lets-workflow/member-run/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/restarter/lets-workflow/member-run"><img src="https://agentmods.dev/badge/skills/restarter/lets-workflow/member-run.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 84 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 2,283 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00084	$0.02283
Opus 5.5	$0.00034	$0.00913
Sonnet 5.5	$0.00017	$0.00457
Haiku 4.5	$0.00008	$0.00228

Measured 2d ago against content hash 89066269992d, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

member-run scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/lets/skills/member-run/SKILL.md · 122 lines

How it starts

The opening of the file, as written. The whole thing — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Member Run

One operation on one named member. The caller (/lets:execute Step 5-D, /lets:team) owns the plan, the gates, the record (the execute run record, the team file), the diff check, the commits and the tracker; this skill owns only the agent call. ListAgents judges who is live; no registry exists beside the caller's record. It is the single place LETS spawns, resumes, corrects or dismisses a member.

IMPORTANT: If the spec below invokes any deferred tool (e.g. AskUserQuestion, SendMessage), you MUST load and call it as specified. Never skip the call, never substitute a default answer of your own - the tool invocation is part of the contract. This is critical.

Args

op	args	does
`spawn`	`scope=<s> name=<n> role=<r> brief-file=<path> report-file=<abs path> [model=<m>] [isolation=worktree]`	Step 1 when it applies, then Step 2
`next`	`scope=<s> name=<n> brief-file=<path> report-file=<abs path> [id=<agent_id>]`	Step 3 - the member's next brief
`correct`	`scope=<s> name=<n> correct-file=<path> report-file=<abs path> [id=<agent_id>]`	Step 3 - an amendment to its current brief
`dismiss`	`scope=<s> name=<n> [id=<agent_id>]`	Step 4

scope - the team's callsign, or run-<RUN> for an execute run.
name - the member name, [a-z0-9-]{1,40}: the bare roster name in a team scope (architect, architect-2), in an execute scope impl-<RUN> (a replacement impl-<RUN>-r<generation>), explorer-<RUN>, skeptic-<RUN>, architect-<RUN>.
{agent} below is the name the agent runs under: <callsign>-<name> in a team scope (pane names are machine-wide), the bare name in an execute scope.
id - the agent_id the caller recorded at spawn; {agent} below is that id when passed, else the name: every SendMessage and TaskStop below puts the id where it says {agent}. An isolated member, which the harness runs as a background agent, is reachable only by its id.
role - a shipped lets:* agent; lets:actor and lets:lead -> stop with role_not_allowed, nothing spawned.
brief-file / correct-file - repo-root-relative paths the caller wrote. Every brief crosses as a file path, never inline.
model - opus | sonnet | fable | haiku, the values the Agent tool accepts. Absent -> the role's own default, except for an implementer (Step 1).
report-file - an absolute path the CALLER named (execute: its pinned .lets/cache/run-... path; team: agent-report op=open / op=add). One file per round; member-run never composes a report path. Required on spawn, next and correct: missing -> stop with report_file_missing, nothing sent.

Read the full file on GitHub · 122 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2d ago Changed 89066269992d
3d ago Changed · -18 lines · +10 tokens per session 4a5e58aaedf8
15d ago Changed · +8 lines fb6028cb4273
20d ago First seen · 132 lines · 74 tokens per session scan A ab9003904f1b

Subscribe to this mod's changes

member-run is a skill published in the GitHub repository restarter/lets-workflow (17 stars, last pushed today), licensed MIT. It adds 84 tokens to every session and 2,283 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-27.

Building agents Claude Code Code review ai-agents claude-code-plugin cli developer-tools

dekko-verify

Sanity-check a suspiciously low or zero call-graph result from dekko (getcallers, getcallees, findusages, impactedtests, unused) before concluding "no callers" or "dead code," or a heritage/throws-provenance result (query supertypes, query subtypes, query throws) that labels something (external) when you expect it to…

aahlijia/dekko · 151 tokens

dekko-review-context

Give a code-review or PR-description flow a structural head start before reading a diff line by line. Trigger when the user asks for a PR description, a pre-merge summary, or invokes a code-review flow (/code-review, /review, or an equivalent) against a diff in a repo with a .dekko/ directory.

aahlijia/dekko · 74 tokens

triage-report

Drain the .review/ report queue. Resolves un-triaged scan/audit reports (producer-agnostic — check-docs-consistency and any other scanner), dispatches the triage-report subagent (Sonnet) one report at a time for per-finding dispositions (promote / patch / dup / needs-investigation / dismiss), then the gateway absorbs…

RockyHong/super-bootstrap · 151 tokens

member-run