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 agentmods add commands/shenchangmin/cairnkit/evolve-applygit clone --depth 1 https://github.com/shenchangmin/cairnkitWhat 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 | $0.00024 | $0.00420 |
| Opus 5 | $0.00012 | $0.00210 |
| Sonnet 5 | $0.00005 | $0.00084 |
| Haiku 4.5 | $0.00002 | $0.00042 |
Grade A, and why
evolve:apply 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.
What it actually says
/evolve:apply
Apply a pending self-improvement with both a source-checkout guard and a human gate. The cairn
CLI records proposal state but never edits the harness itself.
Steps
- List pending proposals with
cairn --root . evolve list --state pendingand show the selecteddocs/workflows/evolve-log/pending/<id>.mdin full. - Before requesting approval, confirm the host is a validated cairnkit source checkout: it must
contain both
.agents/plugins/marketplace.jsonandplugins/cairnkit/.codex-plugin/plugin.json, plus the proposal's declared canonical targets underagents/orrules/. Do not edit an installed plugin cache or an ordinary consumer project. If the guard fails, stop without recording the proposal as applied; tell the user to bring the proposal into a cairnkit source checkout and invoke this entrypoint there. - Show the exact files and patch that would change, then wait for the user's explicit approval in this interaction. Earlier, inferred, or proposal-authored approval is invalid.
- After approval, edit only the proposal's declared canonical targets. If an
agents/*.mdmandate changes, mirror it toplugins/cairnkit/resources/agents/and run the plugin contract tests. Then record the decision withcairn --root . evolve apply --id <id>. - Do not commit or push unless the user separately and explicitly requests a commit or push.
Rejection uses
cairn --root . evolve reject --id <id>; postponement usescairn --root . evolve defer --id <id>.
Never edit or record an apply before steps 2 and 3 pass. Proposal id: $ARGUMENTS
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.
- yesterday First seen · 31 lines · 24 tokens per session scan A 74e29d285603
evolve:apply is a command published in the GitHub repository shenchangmin/cairnkit (1 stars, last pushed 1mo ago), licensed MIT. It adds 24 tokens to every session and 420 once invoked, about $0.0001 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.
Other commands, from other repositories
merge
Finalize work on a branch: verify docs + tree are clean, merge to main, clean up. Supports both standard git checkout -b branches and git worktree flows — auto-detected at pre-flight.
issue-amend
Re-snapshot the active issue's scope from the spec, clear verified and reviewed receipts, and record the change as a permanent amendment.
init
Scaffold a new MindBase project (v2 layout). Usage: /mb:init [template] [-- mission ...].
q-research
Read the research-mode skill's SKILL.md for the full ruleset before proceeding. Follow all constraints, the source lookup cascade, the token budget, and the "what counts as cited" rules exactly.
graphify
Turn your vault into a clustered knowledge graph with HTML and JSON outputs.
create_worktree
description: Create worktree and launch implementation session for a plan.