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.
npx skills add xg-gh-25/SwarmAI --skill s_self-evolutiongit clone --depth 1 https://github.com/xg-gh-25/SwarmAIWrote 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.
[](https://agentmods.dev/skills/xg-gh-25/swarmai/s_self-evolution)<a href="https://agentmods.dev/skills/xg-gh-25/swarmai/s_self-evolution"><img src="https://agentmods.dev/badge/skills/xg-gh-25/swarmai/s_self-evolution/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.
<a href="https://agentmods.dev/skills/xg-gh-25/swarmai/s_self-evolution"><img src="https://agentmods.dev/badge/skills/xg-gh-25/swarmai/s_self-evolution.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00112 | $0.00856 |
| Opus 5 | $0.00056 | $0.00428 |
| Sonnet 5 | $0.00022 | $0.00171 |
| Haiku 4.5 | $0.00011 | $0.00086 |
Grade A, and why
self-evolution 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 9d 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.
How it starts
The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Self-Evolution — Governance Lifecycle Manager
Manages the Three-Layer Governance OS: Principles → Rules → Gates. Read INSTRUCTIONS.md for the full protocol.
Six Operations
| Operation | Trigger | What It Does |
|---|---|---|
| CLASSIFY | Any governance change proposed | Classify layer, parent principle, check conflict/budget |
| CAPTURE | User correction detected | Write C-entry with bias tag (A/B/C/D) |
| PROMOTE | Pattern seen 3x OR user approves | Elevate from EVOLUTION → AGENT/STEERING (absorbs steeringify) |
| REFINE | Principle not precise enough | Sharpen existing principle/rule wording |
| RETIRE | Rule covered by gate, or 30d idle | Remove from active rules, record in EVOLUTION |
| COMPRESS | 3+ rules share same root | Merge into single rule with broader coverage |
Trigger Detection (unchanged from v1)
🔴 Reactive — Capability gap
🟡 Proactive — Better way exists (deferred)
🟢 Correction Capture — User pushback → C-entry with bias tag
🔵 Stuck — Going in circles
Priority: Stuck > Reactive > Correction > Proactive Limits: Max 3 triggers/session. 60s cooldown.
Bias Classification (on every correction)
| Bias | Pattern | Indicator |
|---|---|---|
| A | Premature completion | "Good enough", skip review, defer known issue |
| B | Inference over evidence | Assert without reading, stale memory |
| C | Productivity over quality | Rush output, shallow analysis |
| D | Surrender on obstacle | Give up, ask user to compensate |
Governance Budget (enforced on PROMOTE/REFINE)
- SOUL.md principles: ≤12
- AGENT.md rules: ≤25
- STEERING.md standing rules: ≤15
At cap → COMPRESS or RETIRE existing before adding.
Intake Classification (on every governance change)
Before writing to SOUL/AGENT/STEERING:
- Layer: principle / rule / gate?
- Parent: P1-P7?
- Conflict: duplicates/contradicts existing?
- Budget: current count vs cap
- Source: user / correction / pipeline / self-detect
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 83 lines · 112 tokens per session scan A aa59bcb57981
self-evolution is a skill published in the GitHub repository xg-gh-25/SwarmAI (44 stars, last pushed 2d ago), licensed MIT. It adds 112 tokens to every session and 856 once invoked, about $0.0006 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 2026-08-30.
Other skills, from other repositories
workspace-analyzer
Inspect a workspace, identify temporary files, and perform only approved cleanup or organization steps.
project-notetaker
Formats unstructured meeting notes into a structured project report. Use this when the user provides raw notes or requests a summary of a meeting.
remember
Routes user requests containing "remember", "recall", "checkpoint", "session", "todo", or "where were we" to the correct OpenEmpiric (OEM) MCP tool. Use when the user wants to persist, retrieve, or contextualize knowledge from project memory.
tura
Work in the Tura agent-runtime repository. Use for Tura architecture, Rust backend, GUI/TUI, prompts, commands, providers, sessions, documentation, tests, packaging, and release work in this directory.
compile
Compile a deterministic CIGAR context bundle for the current task with an explicit token budget and inspectable manifest.
why
Explain the provenance, authority, expiry, degradation state, and token accounting of CIGAR context already presented in this session.