Agent skill

Singleton Dispatch

by closedloop-ai in closedloop-ai/claude-plugins

PLN-725 single-agent dispatch for stage11extractsignals and stage15coveragecritic.

Apache-2.0Auto-check passed

Install Singleton Dispatch

skills CLI
$ npx skills add closedloop-ai/claude-plugins --skill singleton-dispatch -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install closedloop-ai/claude-plugins singleton-dispatch --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/closedloop-ai/claude-plugins.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/code-review/skills/singleton-dispatch .claude/skills/singleton-dispatch && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
singleton-dispatch
GitHub stars
122
Token cost
~1.7k tokens
SKILL.md length
702 words
Files
1
Skills in repo
43
Repo updated
First seen
Licence
Apache-2.0

At a glance

PLN-725 single-agent dispatch for stage11extractsignals and stage15coveragecritic.

  • Works in 4 steps: subagent_type:… → model set to the manifest's model field. → Prompt template → …
  • The reviewer fleet (stage20 — see the spawn-reviewers skill)
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
  • The verifier fleet (stage23 — see the verify-findings skill)

What it does

Singleton Dispatch is an agent skill from closedloop-ai/claude-plugins. PLN-725 single-agent dispatch for stage11extractsignals and stage15coveragecritic. After each prepare stage writes its manifest, this skill reads the status field (cachehit / skipped / needsagent) and, only on needsagent, spawns one synchronous singleton Task that writes the by-convention pln725.json output the sibling consolidate stage consumes. Invoke from walker-contract step 6 immediately after stage11extractsignals or stage15coveragecritic finishes. Do NOT use for the reviewer fleet (stage20 — see the…

Its SKILL.md is about 1.7k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The repository describes itself as: Open-source Claude Code plugins for multi-agent software delivery. Plan-first SDLC workflow, code review, LLM quality judges, and self-learning — grounded in your codebase… The licence is Apache-2.0.

When your agent uses it

  • The reviewer fleet (stage20 — see the spawn-reviewers skill)
  • The verifier fleet (stage23 — see the verify-findings skill)

Example prompts

  • “/singleton-dispatch”

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. subagent_type: "code-review:code-review-worker" (same allowlist as the verifier fleet — Read, Write, Grep, Glob).
  2. model set to the manifest's model field.
  3. Prompt template
  4. After the Task returns, hand control back to the caller. In the default run-prefix flow (the caller is start.md's Deterministic Prefix…

What it can do on your machine

Read from SKILL.md and the folder at commit 0e20ac0. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Singleton Dispatch loads about 1.7k tokens when it runs. Until then it costs about 159 tokens; SKILL.md has 702 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~159
When it runs · the whole SKILL.md, loaded when a task matches
~1.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from closedloop-ai/claude-plugins at commit 0e20ac0, republished under its Apache-2.0 licence (© closedloop-ai). 702 words, ~1,728 tokens.

Download SKILL.mdSave it as .claude/skills/singleton-dispatch/SKILL.md (or your agent's skills folder).
name
singleton-dispatch
description
PLN-725 single-agent dispatch for stage_11_extract_signals and stage_15_coverage_critic. After each prepare stage writes its manifest, this skill reads the status field (cache_hit / skipped / needs_agent) and, only on needs_agent, spawns one synchronous singleton Task that writes the by-convention pln725_*.json output the sibling consolidate stage consumes. Invoke from walker-contract step 6 immediately after stage_11_extract_signals or stage_15_coverage_critic finishes. Do NOT use for the reviewer fleet (stage_20 — see the spawn-reviewers skill) or the verifier fleet (stage_23 — see the verify-findings skill).

PLN-725 Single-Agent Dispatch (stage_11 / stage_15)

This skill is the canonical PLN-725 single-agent dispatcher for /code-review. It is split out of commands/start.md so the orchestration spine stays lean; the orchestrator invokes it from walker-contract step 6 immediately after stage_11_extract_signals or stage_15_coverage_critic finishes. The content below is relocated verbatim from start.md.


PLN-725 Single-Agent Dispatch

Walker contract step 6 points here. Two stages (stage_11_extract_signals, stage_15_coverage_critic) emit a manifest whose status field decides whether a singleton LLM agent must be spawned. This section codifies the dispatch — the same protocol applies to both stages with different paths.

When to dispatch

After the prepare stage finishes, read its manifest:

StageManifest location
stage_11_extract_signals<CR_DIR>/extract_signals_manifest.json
stage_15_coverage_critic<CR_DIR>/coverage.json (critic section)

Parse status:

statusAction
"cache_hit"Skip dispatch. Prepare already wrote the canonical output (extract_signals.json for stage_11; coverage.json.final for stage_15). The downstream sibling consolidate stage will no-op.
"skipped"Skip dispatch. Emitted by coverage-critic-prepare for any of three reasons (carried in the manifest's reason field): no-critic (operator passed --no-critic), no-roster (loaded available_reviewers.json is empty — no project agents configured), or no-candidates (roster is non-empty but every reviewer is already in the initial plan — nothing left for the critic to propose). In all three, prepare wrote coverage.json.final with critic_status: "skipped". The downstream sibling consolidate stage will no-op.
"needs_agent"Spawn the singleton agent below.
Spawn contract

For status: "needs_agent", the manifest carries the inputs the agent needs:

FieldUse
prompt_pathThe agent's system prompt path — pass as {PROMPT_PATH} below; the agent reads it itself.
input_pathThe bounded input bundle (taxonomy + diff summary + AVAILABLE list / etc.) — pass as {INPUT_PATH}; the agent reads it itself.
modelModel tier for the dispatch (haiku for signal-extraction, sonnet for coverage-critic).

The manifest's output_path / coverage_state field is the canonical sibling consolidate output (extract_signals.json for stage_11; <CR_DIR>/coverage.json — the final section — for stage_15), NOT where the agent writes. The agent's write target is fixed by convention:

StageAgent writes to
stage_11_extract_signals<CR_DIR>/pln725_extract_signals.json
stage_15_coverage_critic<CR_DIR>/pln725_coverage_critic.json

These paths match the --agent-output arg the sibling stage's args declare, so the consolidate cmd finds the file with no walker substitution.

Spawn one synchronous Task (do not set run_in_background: true). Unlike the reviewer / verifier fleets — which spawn many background tasks and collect them with TaskOutput — the PLN-725 dispatch is a singleton that the walker waits on inline. A synchronous Task returns after the agent finishes; there is no task handle to pass to TaskOutput and no wait step needed:

  1. subagent_type: "code-review:code-review-worker" (same allowlist as the verifier fleet — Read, Write, Grep, Glob).

  2. model set to the manifest's model field.

  3. Prompt template:

    You are the PLN-725 SINGLE-AGENT DISPATCH ({STAGE_LABEL}). Your system
    prompt is at:
      {PROMPT_PATH}
    
    Read it. Your input bundle is at:
      {INPUT_PATH}
    
    Read the bundle for the contract you are validating against. Write
    your output JSON to:
      {OUTPUT_PATH}
    
    Do not write anywhere else. Do not read source files in the
    repository unless the system prompt explicitly says so.

    Placeholder substitution map (do NOT substitute every placeholder from the manifest — output_path in the manifest is the consolidate target, not the agent target):

    PlaceholderSource
    {PROMPT_PATH}manifest.prompt_path
    {INPUT_PATH}manifest.input_path
    {OUTPUT_PATH}The by-convention agent write target from the table above — <CR_DIR>/pln725_extract_signals.json (stage_11) or <CR_DIR>/pln725_coverage_critic.json (stage_15). NOT manifest.output_path.
    {STAGE_LABEL}"signal-extraction" for stage_11, "coverage-critic" for stage_15.
  4. After the Task returns, hand control back to the caller. In the default run-prefix flow (the caller is start.md's Deterministic Prefix loop): do NOT advance the walker — return, and start.md re-invokes run-prefix --resume-from <resume_stage>, which runs the sibling consolidate stage. In the per-stage fallback walk (the caller is walking stages one at a time): advance the walker to the sibling consolidate stage yourself. Either way, no TaskOutput call — that's for background tasks; synchronous Tasks complete before control returns.

Show full SKILL.md (153 more words)Show less
Failure semantics
  • A missing pln725_*.json output after the Task returns is NOT a fatal error. The sibling consolidate stage detects the missing file and fails closed — extract-signals emits the fail-closed default signal set + a signal-extraction-failed finding; coverage-critic leaves the initial plan as final + a coverage-critic-failed finding.
  • Do not retry the singleton agent in the walker. The fail-closed downstream handling is the canonical recovery path — retrying would burn tokens on a request that already failed.
  • Do not parse the agent output in the orchestrator. The sibling consolidate stage is the only reader; orchestrator parsing would duplicate the validation contract.
Cache hits and the skip reasons

Both "cache_hit" and "skipped" (any of no-critic / no-roster / no-candidates) mean prepare wrote the canonical output file itself. The walker advances to the sibling consolidate stage as normal; the sibling stage's cmd detects the manifest status, writes a one-line no-op JSON to stdout, and returns 0. No orchestrator-side branching needed.

© closedloop-ai, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in plugins/code-review/skills/singleton-dispatch of closedloop-ai/claude-plugins.

Open the folder on GitHubat commit 0e20ac0

Compare with similar skills

Singleton Dispatch next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Singleton Dispatch compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Singleton Dispatch this skillclosedloop-ai/claude-plugins122—~1.7kAutomated safety check: PassApache-2.0
Dispatchsickn33/agentic-awesome-skills47k1 repos~2.1kAutomated safety check: PassMIT
Extractalirezarezvani/claude-skills28k—~1.4kAutomated safety check: PassMIT
Matlab Extract Signal Featuresmatlab/matlab-agentic-toolkit1.1k—~4.6kAutomated safety check: PassCustom licence
Brand Extractnexu-io/open-design100k—~3.1kAutomated safety check: PassApache-2.0
Design Extractnexu-io/open-design100k—~549Automated safety check: PassApache-2.0

Similar skills

  • Dispatch

    sickn33/agentic-awesome-skills

    Delegate tasks to OpenAI Codex CLI and Google Antigravity CLI from Claude Code with topic-aware sessions

    47k GitHub starsUsed in 1 repo~2.1k tokens
    Auto-check passed
  • Extract

    alirezarezvani/claude-skills

    Turn a proven pattern or debugging solution into a standalone reusable skill with SKILL.md, reference docs, and examples.

    28k GitHub stars~1.4k tokensUpdated 1 mo ago
    DevelopmentAuto-check passed
  • Matlab Extract Signal Features

    matlab/matlab-agentic-toolkit

    Extract features from 1D signals using signalTimeFeatureExtractor, signalFrequencyFeatureExtractor, and signalTimeFrequencyFeatureExtractor.

    1.1k GitHub stars~4.6k tokensUpdated 7 days ago
    Data & AnalyticsAuto-check passed
  • Brand Extract

    nexu-io/open-design

    Extract a complete Brand Kit from a live website by driving the in-app browser.

    100k GitHub stars~3.1k tokensUpdated today
    Productivity & AutomationAuto-check passed
  • Design Extract

    nexu-io/open-design

    Extract design tokens (color / typography / spacing) from imported source code, screenshots, or Figma exports into the canonical token bag token-map consumes.

    100k GitHub stars~549 tokensUpdated today
    Frontend & DesignAuto-check passed
  • Kg Extract

    ruvnet/ruflo

    Extract entities and relations from source files to build a knowledge graph

    74k GitHub stars~751 tokensUpdated today
    Knowledge ManagementAuto-check: notes

More from closedloop-ai/claude-plugins

All 43 skills in this repo
  • Codex Review

    closedloop-ai/claude-plugins

    Run Codex to review a plan file and return structured feedback with a verdict.

    122 GitHub stars~1.4k tokensUpdated today
    Auto-check: notes
  • Critic Cache

    closedloop-ai/claude-plugins

    Check if critic reviews are still valid before re-running Phase 2.5 critics.

    122 GitHub stars~528 tokensUpdated today
    Auto-check: notes
  • Cross Repo Cache

    closedloop-ai/claude-plugins

    Check if cross-repo coordinator results can be reused, avoiding redundant Sonnet agent launches.

    122 GitHub stars~683 tokensUpdated today
    Auto-check: notes
  • Eval Cache

    closedloop-ai/claude-plugins

    Check for a cached plan-evaluation.json result before launching the plan-evaluator agent.

    122 GitHub stars~516 tokensUpdated today
    Auto-check: notes
  • Find Plugin File

    closedloop-ai/claude-plugins

    This skill should be used when needing to locate files within the Claude Code plugins cache directory (~/.claude/plugins/cache).

    122 GitHub stars~812 tokensUpdated today
    Auto-check passed
  • Gh Monitor PR

    closedloop-ai/claude-plugins

    Start a detached GitHub pull-request monitor that wakes the exact launching Codex Desktop or CLI root through the managed Codex App Server when review, CI, conflict, merge-queue, closure, readiness…

    122 GitHub stars~5.1k tokensUpdated today
    Auto-check passed

Questions about Singleton Dispatch

What does Singleton Dispatch do?

PLN-725 single-agent dispatch for stage11extractsignals and stage15coveragecritic. Singleton Dispatch is an agent skill from closedloop-ai/claude-plugins. PLN-725 single-agent dispatch for stage11extractsignals and stage15coveragecritic.

When should I use Singleton Dispatch?

Singleton Dispatch fits situations like: the reviewer fleet (stage20 — see the spawn-reviewers skill); the verifier fleet (stage23 — see the verify-findings skill).

How do I install Singleton Dispatch in Claude Code?

Run `npx skills add closedloop-ai/claude-plugins --skill singleton-dispatch -a claude-code`. Or copy the skill folder (plugins/code-review/skills/singleton-dispatch in closedloop-ai/claude-plugins) into .claude/skills/singleton-dispatch in your project. Claude Code loads it when a task matches its description.

How do I install Singleton Dispatch in Codex?

Run `npx skills add closedloop-ai/claude-plugins --skill singleton-dispatch -a codex`. Or copy the skill folder (plugins/code-review/skills/singleton-dispatch in closedloop-ai/claude-plugins) into .agents/skills/singleton-dispatch in your project. Codex loads it when a task matches its description.

Can I use Singleton Dispatch in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add closedloop-ai/claude-plugins --skill singleton-dispatch -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/singleton-dispatch, .gemini/skills/singleton-dispatch, .github/skills/singleton-dispatch and .opencode/skills/singleton-dispatch in your project.

What does Singleton Dispatch need to run?

SKILL.md names no scripts, command-line tools or credentials: Singleton Dispatch is instructions for the agent only.

Does Singleton Dispatch access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Singleton Dispatch safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Singleton Dispatch use?

Singleton Dispatch is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Singleton Dispatch use?

About 1.7k tokens (SKILL.md is roughly 6.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Singleton Dispatch?

Skills that share tags, products or a category with Singleton Dispatch: Dispatch (sickn33/agentic-awesome-skills, 47k stars), Extract (alirezarezvani/claude-skills, 28k stars), Matlab Extract Signal Features (matlab/matlab-agentic-toolkit, 1.1k stars) and Brand Extract (nexu-io/open-design, 100k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Singleton Dispatch?

closedloop-ai (a GitHub organization) maintains it in closedloop-ai/claude-plugins, which has 122 GitHub stars. The repository holds 43 skills in this directory. The repository was last updated on October 7, 2026.

Source: closedloop-ai/claude-plugins on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.