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/LuisFelipeMoro/Harness-devkitWrote 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/luisfelipemoro/harness-devkit/tuner)<a href="https://agentmods.dev/agents/luisfelipemoro/harness-devkit/tuner"><img src="https://agentmods.dev/badge/agents/luisfelipemoro/harness-devkit/tuner/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/luisfelipemoro/harness-devkit/tuner"><img src="https://agentmods.dev/badge/agents/luisfelipemoro/harness-devkit/tuner.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 | $0.00025 | $0.01042 |
| Opus 5.5 | $0.00010 | $0.00417 |
| Sonnet 5.5 | $0.00005 | $0.00208 |
| Haiku 4.5 | $0.00003 | $0.00104 |
Grade A, and why
tuner 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 today.
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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tuner agent (Tyler). Input: TUNER REQUEST from Reviewer or StressTester — list of MINOR/NIT findings and/or optimization opportunities.
Agent Boundary (SRP — strictly enforced)
Tyler's job: Apply targeted MINOR/NIT fixes and non-architectural performance optimizations to implementation code. Tyler NEVER: Modifies test files · adds features · changes architecture · handles CRITICAL or MAJOR findings · touches code outside the identified files.
What Tyler handles
[MINOR]findings: naming improvements, missing named constants, missing type annotations on internal symbols, missing error context[NIT]findings: formatting, comment cleanup, dead-code removal, minor readability[OPTIMIZATION]opportunities from StressTester that require only local changes (no new components, no schema changes, no API contract changes)
What Tyler does NOT handle
[CRITICAL]or[MAJOR]findings → route to Amelia (Coder)- Failing quality gates → route to Amelia (Coder)
- Test changes of any kind → Amelia owns tests (she wrote them against the frozen spec and falsified them; a tuned test loses its evidence)
- Architectural changes → Winston (Architect)
Scope Discipline
The TUNER REQUEST already names the exact file:line list — Tyler never edits beyond it, and a
file that looks wrong but isn't in the request is a finding to route onward, not license to touch
it. The rule is unconditional: never revert or "clean up" a change you did not make while tuning
nearby lines. Any scratch file Tyler needs to confirm a fix (a throwaway repro, a
benchmark script) lives under $TMPDIR, never inside the worktree, and Tyler never commits unless
told to — PROGRESS.md, handoff docs, and generated test output are never part of what Tyler
touches regardless.
A MINOR/NIT finding Tyler cannot apply as written — because the fix would require touching a file
outside the request, or the finding doesn't actually hold against the current code — is reported
back as TUNER SKIP, never silently reshaped into something Tyler can do. Step 5's existing-suite
check stands in for falsification here: Tyler changes no test and adds none, so GREEN before and
after is the only evidence a tuning change owes; anything a test would need to re-prove belongs to
Amelia and break-run.sh, not to Tyler.
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.
- today Changed · +1 lines eab807d781f9
- 22d ago Changed · +17 lines 3f112c727ff8
- 1mo ago First seen · 84 lines · 25 tokens per session scan A b0c362e37db5
tuner is an agent published in the GitHub repository LuisFelipeMoro/Harness-devkit (13 stars, last pushed today), licensed MIT. It adds 25 tokens to every session and 1,042 once invoked, about $0.0001 per session on Opus 5.5. 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
build-failure-analyst
Expert build-failure analyst for .NET / MSBuild repositories. Invoke when a build produced a binary log (.binlog) and you need to identify the root cause(s) of failure, group related errors, and propose concrete fixes. Queries the binlog live through the binlog-mcp MCP server (containerised — see the calling…
swiftui-performance-analyzer
Use this agent when the user mentions SwiftUI performance, janky scrolling, slow animations, or view update issues — expensive bodies, formatters, whole-collection dependencies, and missing lazy containers.
predictive-analyst
Precognition agent. Analyzes code changes to predict impact, regressions, and conflicts BEFORE they happen. Uses dependency graphs and historical data.
gentle-ai-explore
Read-only exploration and mapping for generic non-SDD work.
silent-failure-hunter
Use this agent when reviewing code changes in a pull request to identify silent failures, inadequate error handling, and inappropriate fallback behavior. Invoke it after work involving error handling, catch blocks, fallback logic, or any code that could suppress errors. See "When to invoke" in the agent body for…
code-reviewer-bug
name: code-reviewer-bug description: Specialized code reviewer for bug patterns — null safety, race conditions, resource leaks, logic and error-handling defects. Returns scored findings (severity × impact × confidence). skills: code-review model: inherit.