convergence-checker

convergence-checker is a skill for Claude Code, Codex from DojoGenesis/mcp. It costs 59 tokens per session (3,057 once invoked), scanned A, original, MIT.

A project check that measures unfinished work and accumulated changes, then reports RED, YELLOW, or GREEN status. It is intended to show whether development is drifting away from a clean, validated state.

In plain words
What is it for?
Use it before large parallel agent tasks, after warning thresholds are reached, or when you want to check whether the project is ready to reset and continue.
Why use it?
Projects can appear to progress while uncommitted files, open tasks, and repeated work sessions build up. This check makes that drift visible before it becomes expensive to correct.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: model in frontmatter; reads .claude/ paths; mentions CLAUDE.md.

Good fit Use it before large parallel agent tasks, after warning thresholds are reached, or when you want to check whether the project is ready to reset and continue.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/dojogenesis/mcp/convergence-checker
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add DojoGenesis/mcp --skill convergence-checker
Clone the repo
git clone --depth 1 https://github.com/DojoGenesis/mcp

Made for: Claude Code, Codex.

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.

agentmods badge for convergence-checker

README.md
[![agentmods](https://agentmods.dev/badge/skills/dojogenesis/mcp/convergence-checker.svg)](https://agentmods.dev/skills/dojogenesis/mcp/convergence-checker)
Your own site
<a href="https://agentmods.dev/skills/dojogenesis/mcp/convergence-checker"><img src="https://agentmods.dev/badge/skills/dojogenesis/mcp/convergence-checker.svg" alt="Measured on agentmods" height="20"></a>
Per session 59 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,057 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00059 $0.03057
Opus 5 $0.00030 $0.01528
Sonnet 5 $0.00012 $0.00611
Haiku 4.5 $0.00006 $0.00306

Measured 7d ago against content hash 2432692b53b4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

convergence-checker 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 7d 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.

internal/skills/bundled/dojo-craft/convergence-checker/SKILL.md · 280 lines

How it starts

The opening of the file, as written. The whole thing — 280 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Convergence Checker Skill

I. Philosophy

Sonnet sessions produce excellent artifacts but poor trajectory. Each session is locally optimal — clean build, tests pass, feature shipped — but globally drifting: more open items, more uncommitted state, more deferred validation. The missing discipline is periodic evaluation of trajectory, not just output.

Convergence is not a feature session. It produces no new features. It produces a clean, tested, committed, evaluated state — and a reset counter so the next cycle starts from solid ground.

The drift thresholds are intentionally conservative. YELLOW fires at 10 dirty files or 4 sessions — not because the project is in trouble at that point, but because that is when intervention is cheap. RED is expensive. YELLOW is the efficient intervention point.

Do not skip convergence because "things are going well." Drift is invisible until it is expensive.

II. When to Use

  • When drift-detector.sh fires a YELLOW warning (10+ dirty files OR 4+ sessions since last convergence)
  • When drift-detector.sh fires a RED warning (25+ dirty files OR 6+ sessions)
  • Before dispatching a large parallel agent swarm — convergence ensures the baseline is clean
  • After a multi-session sprint to consolidate before the next phase
  • When you sense accumulating open items but cannot articulate the current priority

Do not use this skill for feature work. Convergence sessions produce zero new features. Any fix discovered during convergence is logged and deferred — never implemented in-session.

III. Workflow

Step 1: MEASURE

Establish the current quantitative state.

Run the following:

# Count uncommitted changes per repo
git status --short | wc -l

# List all uncommitted files
git status --short

# Count untracked files separately
git status --short | grep "^?" | wc -l

# Check for sub-repos with dirty state
find . -name ".git" -maxdepth 3 -type d | while read gitdir; do
  repo=$(dirname "$gitdir")
  count=$(git -C "$repo" status --short 2>/dev/null | wc -l | tr -d ' ')
  if [ "$count" -gt "0" ]; then
    echo "$repo: $count dirty files"
  fi
done

Read the full file on GitHub · 280 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.

  1. 7d ago First seen · 280 lines · 59 tokens per session scan A 2432692b53b4

Subscribe to this mod's changes

convergence-checker is a skill published in the GitHub repository DojoGenesis/mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 59 tokens to every session and 3,057 once invoked, about $0.0003 per session on Opus 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 2026-08-31.

Related

Other skills, from other repositories