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.
git clone --depth 1 https://github.com/rjwalters/kicad-toolsWrote 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/commands/rjwalters/kicad-tools/champion-epic-guard-invariants)<a href="https://agentmods.dev/commands/rjwalters/kicad-tools/champion-epic-guard-invariants"><img src="https://agentmods.dev/badge/commands/rjwalters/kicad-tools/champion-epic-guard-invariants/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/commands/rjwalters/kicad-tools/champion-epic-guard-invariants"><img src="https://agentmods.dev/badge/commands/rjwalters/kicad-tools/champion-epic-guard-invariants.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00000 | $0.02344 |
| Opus 5.5 | $0.00000 | $0.00938 |
| Sonnet 5.5 | $0.00000 | $0.00469 |
| Haiku 4.5 | $0.00000 | $0.00234 |
Grade A, and why
champion-epic-guard-invariants 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 — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Champion: Idempotency Guard for Unrevised Epics — Maintainer Notes
Editor-facing notes for champion-epic.md → "Idempotency Guard for Unrevised
Epics": why the marker keys on a body hash, what each counter counts, the
cycle-by-cycle trace, and the invariants a future edit must preserve. A running
Champion pass needs nothing here — the guard's own code and outcome table are
complete without it. Split out so champion-epic.md stays under its
markdown-token ratchet (#7725), the same move #7666 made for Step 0.5. Read this
before editing the guard or Step 4.
Why a hash of title + body, and NOT the epic's updatedAt
updatedAt is self-invalidating — posting the verdict comment bumps it, so
the marker could never match and every pass would re-comment. A title + body hash
changes if and only if the epic is actually edited. Full derivation:
champion-issue-promo.md → "Why a body hash and NOT the issue's updatedAt
(#4966)".
The counters, and why the skip must cost something
| Mechanism | Counts | Written by | Survives a silent skip? |
|---|---|---|---|
PRIOR_REJECTIONS |
posted Champion Review: Epic Needs Revision comments (any revision) |
Step 4's reject branch | Yes, but frozen while skipping — it cannot advance on its own |
SKIP_STREAK |
silent skips recorded for the current body hash | the skip path's in-place PATCH of the existing verdict comment |
Yes — this is the counter that keeps advancing |
UNREVISED_EVALS = PRIOR_REJECTIONS + SKIP_STREAK |
evaluation cycles spent on an unrevised epic | derived | Yes — the single escalation gate, used identically by the skip path and Step 4 |
Suppressing duplicate comments must never suppress the escalation that puts a stuck epic in front of a human (#4967). Traced against an epic that fails at body hash H1 and is never revised:
| Cycle | Marker match? | PRIOR_REJECTIONS |
SKIP_STREAK |
UNREVISED_EVALS |
Outcome | Comments posted |
|---|---|---|---|---|---|---|
| 1 | no (H1 unseen) | 0 | 0 | 0 | evaluate → reject → post "Epic Needs Revision" carrying VERDICT_MARKER + epic-unrevised-skips:H1:0 |
1 |
| 2 | yes (H1) | 1 | 0 | 1 < 2 | silent skip; PATCH the tally to 1 |
0 |
| 3 | yes (H1) | 1 | 1 | 2 ≥ 2 | ESCALATE_UNREVISED=yes → Step 4 escalation → loom:operator |
1 (escalation) |
| 4+ | — | — | — | — | ALREADY_ROUTED=yes drops it from every future pass |
0 |
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 f0c3bebab6b8
- 21d ago Changed · +11 lines b43dee733ecf
- 24d ago Changed · +10 lines 7db2dd559b2e
- 29d ago Changed · +18 lines 5a9299f6ec71
- 1mo ago First seen · 85 lines · 0 tokens per session scan A 6b7b821e54a5
champion-epic-guard-invariants is a command published in the GitHub repository rjwalters/kicad-tools (70 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,344 tokens. 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 2026-09-17.
Other commands, from other repositories
template
Manage issue templates for streamlined issue creation.
fest-show
Show festival progression (in-progress tasks, roadmap, and dependency view).
add-note
Add an internal or external note to a ConnectWise PSA ticket.
sync-linear
Sync current work with Linear ticket status.
dispatcher
Use when you want the orchestrator to pick the next repo to work on across your whole portfolio — it enumerates candidate repos below the confinement root, resolves free/busy from each repo's session.lock lease, ranks the FREE ones by backlog priority × staleness × readiness, recommends the single most worthwhile one…
workpm
A project-management workflow for coordinating multiple AI workers through five stages. It includes task assignment, shared activity logs, worker replacement, and final checks.