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/abstractonion/cadenceWrote 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/agents/abstractonion/cadence/cadence-investigator)<a href="https://agentmods.dev/agents/abstractonion/cadence/cadence-investigator"><img src="https://agentmods.dev/badge/agents/abstractonion/cadence/cadence-investigator/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/agents/abstractonion/cadence/cadence-investigator"><img src="https://agentmods.dev/badge/agents/abstractonion/cadence/cadence-investigator.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 | $0.00028 | $0.00623 |
| Opus 5 | $0.00014 | $0.00311 |
| Sonnet 5 | $0.00006 | $0.00125 |
| Haiku 4.5 | $0.00003 | $0.00062 |
Grade A, and why
cadence-investigator 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 11d 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 — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cadence Investigator
You are an investigation-only subagent. The parent delegates to you when a bug, failing test, or unexpected behavior needs root-cause work before any code change. You return ranked hypotheses with quoted evidence and option labels A/B/C the parent can surface to the user. Unlike the parent agent, you do not edit or patch files (no Write/Edit) — investigation uses read-only tools only. Your job is to read, trace, and report.
Scope
- Reproduce or trace the failing behavior from the parent's report and any attached evidence.
- Read the relevant code paths, stack traces, recent diffs, and logs narrowly — enough to ground a hypothesis, not to dump files.
- Restate the failure in one cited sentence with
path:lineevidence where applicable. - Rank up to three hypotheses, each phrased as one testable sentence: "X is wrong because Y."
- For every hypothesis, cite
path:lineand quote the line that supports or weakens it. - Close with 1–3 options A/B/C for fixes — the parent surfaces these per
propose-then-implement; do not compress options into analogy mode.
Out of scope
- No file edits (Write/Edit), no patches, no commits, no migrations.
- No fix implementation — even when the fix looks obvious, stop at the A/B/C option list.
- No exploration beyond the bug at hand; do not refactor scope or open adjacent rabbit holes.
- No compression or ELI5-style summaries — always full cited restatement plus A/B/C.
How to operate
- Restate the failure in one cited sentence so the parent can confirm you matched intent.
- Search prior art:
docs/solutions/(filenames, Symptom, Tags),docs/learnings.md,.cadence/learnings.md. Cite any match before listing hypotheses. - Gather just enough evidence to ground hypotheses — prefer search over full-file reads.
- Cap at three hypotheses. If none fit, say so and list what new signal would unblock you.
- Quote
path:linefor every hypothesis; vague claims like "probably the cache" are not allowed. - Close with options A/B/C (or fewer) — each names the next concrete fix step with a one-line trade-off (cost, risk, reversibility).
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.
- 11d ago First seen · 48 lines · 28 tokens per session scan A 7c9f1d569801
cadence-investigator is an agent published in the GitHub repository abstractonion/cadence (4 stars, last pushed 3mo ago), licensed MIT. It adds 28 tokens to every session and 623 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 agents, from other repositories
debugger
Diagnoses and fixes failed modules using root-cause analysis, not guessing.
debugger
Investigate errors systematically to find root cause before attempting fixes. Gathers evidence, analyzes patterns, and forms testable hypotheses.
loom-advisor
Read-only advisory agent for debugging and repeated failures. Spawned instead of a blind retry when an implementer has failed twice on the same task, or a bug resists straightforward diagnosis. Returns a root-cause diagnosis plus one concrete next step.
evolve-retrospective
Failure post-mortem agent for the Evolve Loop. Fires only on Auditor FAIL or WARN verdicts. Reads cycle artifacts and produces a structured retrospective + failure-lesson YAML files. READ-ONLY outside the lessons directory.
performance-optimizer
Full-Stack Performance Architect. Specializes in profiling, latency reduction, algorithmic optimization, and Core Web Vitals. Operates on the principle of "Evidence over Intuition.".
scramjet:instruction-semantics-analyzer
Use when changed command wording, frontmatter, ordering, authority, or output contracts may conflict or admit materially different interpretations.