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-savantgit 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-savant)<a href="https://agentmods.dev/skills/bop-clocktower/canary/canary-savant"><img src="https://agentmods.dev/badge/skills/bop-clocktower/canary/canary-savant/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/skills/bop-clocktower/canary/canary-savant"><img src="https://agentmods.dev/badge/skills/bop-clocktower/canary/canary-savant.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 | $0.00114 | $0.03115 |
| Opus 5 | $0.00057 | $0.01558 |
| Sonnet 5 | $0.00023 | $0.00623 |
| Haiku 4.5 | $0.00011 | $0.00312 |
Grade A, and why
canary-savant 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 5d 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 — 234 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Canary Savant
A test that only passes because of the tests that ran before it is a lie that
passes CI. Savant finds shared-state leakage and names the culprit. The static
pass is the cheap half that runs on every PR and points at the suspects. The
confirming pass (--confirm, opt-in) proves a leak by shuffling the suite
under a pinned seed and bisecting the prefix to name the polluter — not just the
victim.
Tier-0 by the definition in
ADR 0015:
deterministic, no network, no agent or LLM. Also no secrets and no dependency on
any other skill. Both passes qualify — --confirm runs the suite locally, which
needs no network and no model.
Rules (static pass — suspects)
| Rule | Severity | Fires on |
|---|---|---|
SV001-module-mutable-global |
medium | A module-scope mutable (= {}, = [], set(), dict(), list(), or a top-level JS let/var/const object/array) that some line later mutates in place (.append/.add/[...] =/+=/.attr =). Fires on the declaration, the leak's source. |
SV002-missing-teardown |
medium | A class/all-scoped setup whose matching teardown is absent: pytest setup_class/setUpClass, or vitest/jest beforeAll. Per-test setup (setUp/setup_method/beforeEach) is excluded - it rebuilds state each test, so it does not leak. |
SV003-shared-singleton-mutation |
low | A process-global singleton assigned without restore: os.environ[...] =, sys.modules[...] =, process.env.X =. Reads and == comparisons never fire, and neither does a file that demonstrably restores the global — a same-key (or computed-loop) restore in afterEach/afterAll/teardown/post-yield fixture code or an in-test try/finally, or a write-back from a snapshot saved from that global. |
SV004-order-coupled-name |
low | A test name or comment that encodes ordering: test_1_…, a terminal ordinal (test_first(), test_last() - not test_first_match_wins), must run before …, it('… run first'). |
What ships with it
6 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.
- 5d ago Changed · +4 lines · -4 tokens per session f102e8aa7eb2
- 10d ago First seen · 230 lines · 118 tokens per session scan A ff87ec55aa94
canary-savant is a skill published in the GitHub repository bop-clocktower/canary (4 stars, last pushed today), licensed MIT. It adds 114 tokens to every session and 3,115 once invoked, about $0.0006 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.