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 pingchesu/hermes-curator-evolver --skill curator-evolutiongit clone --depth 1 https://github.com/pingchesu/hermes-curator-evolverWrote 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/pingchesu/hermes-curator-evolver/curator-evolution)<a href="https://agentmods.dev/skills/pingchesu/hermes-curator-evolver/curator-evolution"><img src="https://agentmods.dev/badge/skills/pingchesu/hermes-curator-evolver/curator-evolution.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00041 | $0.00965 |
| Opus 5 | $0.00020 | $0.00483 |
| Sonnet 5 | $0.00008 | $0.00193 |
| Haiku 4.5 | $0.00004 | $0.00097 |
Grade A, and why
curator-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 2d 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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Curator Evolution
Hermes Curator Evolver starts from evidence and keeps mutation guarded. Reports and proposals are review artifacts. auto-run can make skills actually improve, but only through low-risk bounded managed blocks, reference spillover for bulky evidence, and guarded apply.
Interpretation Checklist
- Separate evidence from conclusions.
- Repeated tool errors suggest a possible missing pitfall or verification step in a skill.
- Repeated skill reads suggest the skill is active and worth keeping discoverable.
- A single failure is not enough evidence to rewrite a skill.
- If the skill was correct but the agent ignored it, improve triggers/descriptions only when evidence repeats.
- Candidate search is advisory; embedding/reranker models only find candidates and do not decide edits.
backfill-sessionsreads current Hermesstate.dbsessions through the read-onlySessionDBAPI and retains explicit legacysession_*.jsonsupport; it is model-free and does not mutate skills.auto-run --semantic-candidatesand--rerank-candidatesare explicit opt-ins that only reorder evidence-eligible candidates.- Guarded apply requires approval, backup, verifier/validation pass, and rollback.
auto-runmutates only when both--apply-low-riskand--approve-auto-applyare set.- Even with write flags, unattended auto-apply writes only local agent-created skills. Official/bundled skills (
.bundled_manifest), hub-installed skills (.hub/lock.json), plugin-provided skills,skills.external_dirs, pinned skills, and unknown sources are skipped. - Core-name allowlists only operate inside that local agent-created source boundary; they do not override provenance.
- Size guardrails target a 90k
SKILL.mdsoft cap, spill bulky autorun evidence intoreferences/, and skip unattended writes when the target skill is already over the 100k hard cap. install-auto --enablecreates a native user scheduler: systemd user timer on Linux, launchd LaunchAgent on macOS; remove it withuninstall-autobefore plugin uninstall.
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.
- 2d ago Changed 21be8e51dc40
- 7d ago First seen · 47 lines · 41 tokens per session scan A f259e871311a
curator-evolution is a skill published in the GitHub repository pingchesu/hermes-curator-evolver (38 stars, last pushed 6d ago), licensed MIT. It adds 41 tokens to every session and 965 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 2026-08-30.
Other skills, from other repositories
agent-communication-protocol
Open protocol for AI agent interoperability enabling standardized communication between agents, applications, and humans across different frameworks.
playwright-e2e-testing
Author and maintain versioned Playwright (@playwright/test) TypeScript UI specs for browser user flows. Use when asked to create, run, debug, or refactor E2E tests, form/navigation/auth flows, responsive checks, UI mocking, fixtures, Page Objects, or visual comparisons. Use api-testing for standalone REST/GraphQL…
webapp-selenium-testing
Author and maintain versioned Selenium WebDriver tests with Java and JUnit 5. Use for creating, debugging, or running Selenium specs, implementing Page Objects, handling explicit waits, capturing screenshots, or setting up Maven test projects. Supports Chrome, Firefox, and Edge. Keywords: Selenium WebDriver, Java…
qa-investigation
Investigate a specific test failure to its root cause and document the why. Detects whether a failing test is flaky (intermittent) or a deterministic bug during reproduction. Use when a test fails and you need the real cause, not just to make it green. Execution layer, not strategy review. Keywords: flaky test…
api-integration-test
Create, maintain, and run gated Go integration tests for internal APIs and service-to-service clients (HTTP/gRPC). Use for endpoint verification, contract checks with real runtime config, opt-in execution, timeout/retry safety, and integration failure triage in Go services.
go-dependency-audit
Go dependency audit specialist for CVE scanning (govulncheck), license risk triage, outdated dependency detection, upgrade impact analysis, and supply chain security. ALWAYS use when auditing go.mod dependencies, running govulncheck, checking license compatibility, planning dependency upgrades, or investigating supply…