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/evolvegit 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 | $0.00022 | $0.00226 |
| Opus 5 | $0.00011 | $0.00113 |
| Sonnet 5 | $0.00004 | $0.00045 |
| Haiku 4.5 | $0.00002 | $0.00023 |
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.
What it actually says
/evolve
Diagnose a problem and propose a harness improvement. This never changes the harness — it only writes a proposal for later human review.
Steps
- Replay the relevant workflow log/artifacts under
docs/workflows/<run-id>/and do root-cause analysis. - Simulate the improved behavior in context (what would the agent/rule have needed?).
- Write the analysis + concrete suggested change, then record it:
cairn --root . evolve propose --id <slug> --file <analysis.md>
Do not send a notification or any other external message unless the user separately identifies the target and explicitly requests it in this interaction. Recording the local proposal does not authorize a webhook or channel message.
The proposal lands in docs/workflows/evolve-log/pending/. Apply it only via the guarded
evolve-apply entrypoint.
The issue to analyze: $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 · 25 lines · 22 tokens per session scan A 3d6d7ead7de6
evolve is a command published in the GitHub repository shenchangmin/cairnkit (1 stars, last pushed 1mo ago), licensed MIT. It adds 22 tokens to every session and 226 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
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.
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.
ingest-l1
L1 analysis loop for the abapwiki knowledge base: for each batch it launches the abap-analyzer sub-agent in parallel, then the adversarial judge abap-deepcheck (separate session), applies only the analyses that pass the fail-closed gate, and commits. Resumes exactly after an interruption. Use this skill to document…
export-project
Export a single registered Gnosys project to a portable .json.gz bundle (round-trips with gnosys import project).
ambiguity
Check if a query matches memories in multiple projects.
reindex
Rebuild semantic embeddings for every memory in configured stores.