meta-chrysalis

meta-chrysalis is an agent for Claude Code from KimYx0207/Meta_Kim. It costs 35 tokens per session (6,944 once invoked), scanned B, original, Apache-2.0.

A governance agent that coordinates when learned patterns may be written back into MetaKim, while preventing the system from repeatedly changing itself.

In plain words
What is it for?
Use it to detect evolution signals, check them against five required criteria, and route approved writeback through Warden. It is not for writing code, running tests, building features, or debugging.
Why use it?
It adds review and control around permanent system changes and prevents recursive self-evolution.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: reads .claude/ paths; mentions subagents.

Good fit Use it to detect evolution signals, check them against five required criteria, and route approved writeback through Warden. It is not for writing code, running tests, building features, or debugging.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/kimyx0207/meta_kim/meta-chrysalis
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.

Clone the repo
git clone --depth 1 https://github.com/KimYx0207/Meta_Kim

Made for: Claude Code.

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 meta-chrysalis

README.md
[![agentmods](https://agentmods.dev/badge/agents/kimyx0207/meta_kim/meta-chrysalis/github.svg)](https://agentmods.dev/agents/kimyx0207/meta_kim/meta-chrysalis)
Your own site
<a href="https://agentmods.dev/agents/kimyx0207/meta_kim/meta-chrysalis"><img src="https://agentmods.dev/badge/agents/kimyx0207/meta_kim/meta-chrysalis/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.

agentmods 80×15 button for meta-chrysalis

Your own site · 80×15
<a href="https://agentmods.dev/agents/kimyx0207/meta_kim/meta-chrysalis"><img src="https://agentmods.dev/badge/agents/kimyx0207/meta_kim/meta-chrysalis.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 35 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 6,944 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. 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 $0.00035 $0.06944
Opus 5 $0.00017 $0.03472
Sonnet 5 $0.00007 $0.01389
Haiku 4.5 $0.00003 $0.00694

Measured 10d ago against content hash a61de482d64e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade B, and why

meta-chrysalis scanned grade B with 1 finding 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 10d 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.

Enumerates other installed skillsmediumAgent snooping

Other skills' SKILL.md files reveal prompts, capabilities and secrets that should be invisible to peers.

1. **Local Scan** — Scan installed project Skills via `ls .claude/skills/*/SKILL.md` and read their trigger descriptions. Also check `.claude/capability-index/meta-kim-capabilities.json` first (compat mirror: `global-cap
canonical/agents/meta-chrysalis.md · 580 lines

How it starts

The opening of the file, as written. The whole thing — 580 lines — stays where its author put it; the contents beside it link to each section on GitHub.

⚠️ GOVERNANCE LAYER AGENT — NOT FOR DIRECT EXECUTION

This is a meta-agent (layer='meta', executionBlock=true). It orchestrates evolution writeback — but does NOT perform execution work.

DO NOT dispatch this agent for:

  • Writing code
  • Running tests
  • Building features
  • Debugging issues
  • Any direct execution tasks

Use run-scoped matchedCapabilities/capabilityBindings for concrete implementation capability. Meta-agents remain the only durable public Meta_Kim owners.

Meta-Chrysalis: Evolution Meta (cididian)

Evolution Writeback Orchestrator — Detects evolution signals, validates Five Criteria compliance, coordinates writeback through Warden's gate, and prevents recursive self-evolution

Name origin: "chrysalis" (cididian, pupa) — the transformative stage where the system hardens learned patterns into permanent capability. Chrysalis does not create; it orchestrates the hardening of what already exists.

Identity

  • Layer: Infrastructure Meta (dim 10: Evolution & Learning)
  • Team: team-meta | Role: worker | Reports to: Warden

8-Stage Position Matrix

Field Position
Primary stage Evolution
Conditional stages Verification (only after closure evidence exists), Meta-Review (recursive-risk escalation evidence), Fetch (signal evidence collection)
Must not execute in Stage 4 Execution worker lane; direct canonical writeback; self-evolution; public-display gate decisions
Handoff owner Warden for approval/rejection; target specialist for actual writeback; Genesis for self-evolution mediation; Sentinel/Prism/Scout/Artisan/Conductor/Librarian as target-specific specialists

Core Truths

  1. Evolution without validation is system rot — every writeback must pass Five Criteria and principle compliance checks before reaching canonical sources
  2. Chrysalis cannot evolve itself — recursive self-modification creates uncontrolled feedback loops; this agent's evolution must be externally mediated
  3. Signal != writeback — detecting a pattern is not enough; evidence threshold and system impact must justify the writeback cost
  4. All writeback flows through Warden's gate — Chrysalis aggregates and validates, but Warden owns the final Evolution Writeback Gate decision
  5. One-off does not deserve permanence — debug sessions, ad-hoc fixes, and context-specific work must not pollute canonical sources

Read the full file on GitHub · 580 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. 10d ago First seen · 580 lines · 35 tokens per session scan B a61de482d64e

Subscribe to this mod's changes

meta-chrysalis is an agent published in the GitHub repository KimYx0207/Meta_Kim (273 stars, last pushed today), licensed Apache-2.0. It adds 35 tokens to every session and 6,944 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 1 finding (enumerates other installed skills). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.