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 bop-clocktower/canary --skill canary-shadowgit clone --depth 1 https://github.com/bop-clocktower/canaryWrote 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/bop-clocktower/canary/canary-shadow)<a href="https://agentmods.dev/skills/bop-clocktower/canary/canary-shadow"><img src="https://agentmods.dev/badge/skills/bop-clocktower/canary/canary-shadow.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 | $0.00142 | $0.01516 |
| Opus 5 | $0.00071 | $0.00758 |
| Sonnet 5 | $0.00028 | $0.00303 |
| Haiku 4.5 | $0.00014 | $0.00152 |
Grade A, and why
canary-shadow 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 — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Canary: Shadow (differential parity testing)
Run one invocation through two implementations, normalize the noise, diff the rest. Every un-accepted divergence is a behavior change to explain or fix.
When to Use
- Ports / rewrites where output should be identical: Python→TS, a BoP skill
re-port, typer→commander, one engine replacing another. (This is exactly how
the v6 cutover was validated: run
canary <cmd>through the TS engine AND the Python reference, diff. It surfaced a divergence the unit tests + golden suite missed — Python'srichconsole renders the ticket-marker hint# canary:ticket:as# canary🎫via emoji-shortcode substitution, mangling a hint the parser can't match; the TS port printed the literal, correct marker. A console-rendering artifact only a live side-by-side run exposes.) - Skill changes — old skill vs new skill on the same inputs: "prove the rewrite is equivalent."
- Refactors asserted to be behavior-preserving.
- Framework / dependency swaps where the observable surface should not move.
- NOT for greenfield behavior (nothing to compare against) — use asserted tests.
- NOT as a sole correctness check — see the caveat at the bottom.
The idea (differential / shadow testing)
For each case, run the SAME arguments through a baseline command and a
candidate command, capture {exitCode, stdout, stderr}, normalize both,
and diff. Identical (post-normalization) ⇒ ok. Different ⇒ DIVERGE — a real
behavior change, unless it's a documented, intentional difference recorded in
the accepted-divergence allowlist.
The value is not the runner (two spawns and a diff) — it's the two disciplines
below.
1. The normalization ruleset (get this right, or drown)
Output has meaningful content and irrelevant noise. Normalize the noise; keep
everything else. Default masks (see scripts/cli.mjs): ANSI SGR codes, ISO
timestamps, temp paths (/tmp, /var/folders, /private), version banners
(vX.Y.Z[-rc.N]), commit SHAs, run-ids.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · +17 lines c8cf56440774
- 7d ago First seen · 115 lines · 142 tokens per session scan A ff02f3049312
canary-shadow is a skill published in the GitHub repository bop-clocktower/canary (4 stars, last pushed today), licensed MIT. It adds 142 tokens to every session and 1,516 once invoked, about $0.0007 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 skills, from other repositories
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
neuron-evaluation-engineer
Create and run AI evaluations with datasets, assertions, and output drivers in Neuron AI. Use this skill whenever the user mentions evaluation, testing AI systems, creating evaluators, dataset-driven testing, assertion-based validation, or wants to measure AI system performance. Also trigger for tasks involving…
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.
skill-benchmark
Benchmark AI skill effectiveness by measuring implementation quality against legacy constraints.