evolve

A self-maintenance command that evolves one Guild skill or other supported instance through a promotion process. Promotion means accepting a new version only after it passes the required checks.

In plain words
What is it for?
Use it to evolve a named skill or instance, either interactively or with unattended processing through the optional --auto flag.
Why use it?
It provides a controlled way to improve a skill while keeping rejected attempts archived and preventing unapproved automatic changes.

Command

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.

agentmods
npx agentmods add commands/lookatitude/guild/evolve
Clone the repo
git clone --depth 1 https://github.com/lookatitude/guild
Per session 36 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 710 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00036 $0.00710
Opus 5 $0.00018 $0.00355
Sonnet 5 $0.00007 $0.00142
Haiku 4.5 $0.00004 $0.00071

Measured yesterday against content hash 88cd4ed08553, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

evolve 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 yesterday.

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.

commands/evolve.md · 75 lines

How it starts

The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/guild:evolve — self-maintenance (Guild-on-Guild)

Kicks off skill evolution. Maps to skill guild:evolve-skill.

Canonical grammar

/guild:evolve <id> [--auto]
  • <id> (positional, required) — the skill or evolvable instance to evolve. This is the same single positional the legacy [skill] form named; it is widened, not duplicated<id> covers a base skill name and an instance id.
  • --auto — run the promotion pipeline unattended; the promotion gate is still respected (gate criteria must pass, no regression), unchanged.

Gates

Promotion gate I (manual) / A (--auto, gate-respected). Never auto-edits a skill without passing the promotion gate; rejected attempts are archived, not deleted.

Output

.guild/evolve/<run-id>/**, version bump on promote.

Run-start preflight (settings-control-and-tmux U3/U6)

Before the evolve-skill pipeline runs — and before run-trace start — the run-trace CLI runs this preflight for you — you do not call runStartPreflight yourself.

Since wave 2, run-trace.js start (below) is the sole caller of runStartPreflight (scripts/lib/runstart-preflight.ts; canonical contract in guild.md §Run-start preflight): on start the CLI resolves the 7-source inheritance chain, validates closed keys, probes tmux, detects providers, and writes .guild/runs/<id>/resolved-settings.json (+ a compact settings_ref in run.yaml) automatically before the run opens. If this command needs the resolved config — e.g. the dispatch backend effective.agent_mode — read the snapshot back with readResolvedSettingsSnapshot(runId, { cwd }); never re-resolve.

Run recording

Before the evolve-skill is invoked, start a run (SC-B, §435):

node ${GUILD_PLUGIN_ROOT:-${CLAUDE_PLUGIN_ROOT:-$HOME/.local/share/guild/dist/claude-code}}/hooks/dist/run-trace.js start \
  --command=/guild:evolve \
  --cwd "$(pwd)"

run-class default (full). Records the run before the promotion gate so the complete session — shadow evaluation and promotion decision — is replayable from the entrypoint. Writes to .guild/evolve/ — not .guild/initiatives/ (NN#5 unaffected). No --initiative flag.

Read the full file on GitHub · 75 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. yesterday First seen · 75 lines · 36 tokens per session scan A 88cd4ed08553

Subscribe to this mod's changes

evolve is a command published in the GitHub repository lookatitude/guild (7 stars, last pushed yesterday), licensed MIT. It adds 36 tokens to every session and 710 once invoked, about $0.0002 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.