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 manishiitg/coding-agent-loop --skill performance-regression-to-owned-changegit clone --depth 1 https://github.com/manishiitg/coding-agent-loopWrote 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/manishiitg/coding-agent-loop/performance-regression-to-owned-change)<a href="https://agentmods.dev/skills/manishiitg/coding-agent-loop/performance-regression-to-owned-change"><img src="https://agentmods.dev/badge/skills/manishiitg/coding-agent-loop/performance-regression-to-owned-change/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/manishiitg/coding-agent-loop/performance-regression-to-owned-change"><img src="https://agentmods.dev/badge/skills/manishiitg/coding-agent-loop/performance-regression-to-owned-change.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 | $0.00030 | $0.00581 |
| Opus 5.5 | $0.00012 | $0.00232 |
| Sonnet 5.5 | $0.00006 | $0.00116 |
| Haiku 4.5 | $0.00003 | $0.00058 |
Grade A, and why
performance-regression-to-owned-change 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 19d 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.
Performance Regression to Owned Change
Outcome
Produce a sourced regression diagnosis and an owned blocker ledger with reproduction, uncertainty and next verification.
When to use
Use this route when the job crosses the listed owners or access boundaries. A single Crew may do a compatible read-only task. This package is a Builder proposal, not an active automation.
Discovery and user direction
Inspect existing Crews, source access, policy revisions, owner decisions and the exact subject. Ask only for missing facts, then show the proposed route for review.
Required inputs
One service or page, endpoint or journey, baseline and regressed builds, environment, traffic condition and Engineering owner. Use comparable trace or test runs, performance budget revision, deployment timeline, issue/PR/CI state and change policy.
Plan and AgentWorks tools
Create reviewed steps for performance-investigator (performance-regression-finding/v1) → engineering-delivery-coordinator (engineering-performance-action/v1). Reuse Crews where compatible; Builder may propose create_crew for a missing role. Validate each typed artifact before the next step and independently re-read current source state. No production change, rollback, issue write, customer claim or performance recovery claim from installation or a proposed fix.
Knowledge and persistence
Store tenant, subject, Crew, artifact, source revision, decision, approval and provider receipt IDs. Re-read source truth and deduplicate on repeat; keep prior decisions in the ledger.
Validation and reporting
The contract validator checks exact identity, revisions, time order and route-specific false claims. The first artifact, last artifact and rejected artifact exercise it. The dashboard must separate pending review, owner approval and observed provider action.
Guardrails
A metric change needs comparable conditions and coverage; correlation with a deploy is a hypothesis. An accepted change plan is not a deployed fix. No production change, rollback, issue write, customer claim or performance recovery claim from installation or a proposed fix. Do not count a draft, approval or planned action as a provider outcome.
What ships with it
9 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.
- examples/delivery.json 832 B
- examples/invalid-delivery.json 830 B
- examples/investigation.json 780 B
- playbook.json 5.1 KB
- references/team-and-handoffs.md 1.7 KB
- scripts/additional_route_contract.py 8.8 KB runs code
- scripts/test_validate_handoff.py 1.3 KB runs code
- scripts/validate_handoff.py 649 B runs code
- SETUP.json 3.1 KB
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.
- 19d ago First seen · 48 lines · 30 tokens per session scan A 84358abd832f
performance-regression-to-owned-change is a skill published in the GitHub repository manishiitg/coding-agent-loop (13 stars, last pushed 19d ago), licensed MIT. It adds 30 tokens to every session and 581 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 2026-09-28.
Other skills, from other repositories
ci-troubleshoot
Diagnose failed GitHub Actions runs for pi-agent-dashboard: the 11-file workflow taxonomy, affected-test selection, the release pipeline, known failure modes, and how to read gh run logs and retrigger jobs. Use when a CI run is red, a release is stuck, a workflow won't dispatch, or you need to know which workflow does…
performance-optimization
Measure-first performance work. Use on triggers like "it's slow", "profile this", "optimize perf", "fix the bottleneck", "improve load time / Core Web Vitals", or when a measured regression needs fixing. Enforces measure-before-optimize. Fills a perf gap not covered by existing project skills. Not a build/ship…
node-inspect-debugger
See runtime state a console.log cannot reach — set real breakpoints, step, and dump the scope chain of a paused Node/TypeScript process. Use on "set a breakpoint", "inspect runtime state", "console.log isn't enough", "step through this", "what's in this closure at runtime", "attach a debugger". Not a logging or…
debug-dashboard
Diagnose problems in the running pi-agent-dashboard system: server.log, /api/health, bridge WebSocket connectivity, vitest triage, known-issue FAQ entries. Routes UI/visual issues to the browser skill. Use when the server seems hung, a pi session won't connect, tests fail mysteriously, the dashboard shows a blank…
systematic-debugging
Root-cause a bug already in front of you, instead of guessing at fixes. Use on triggers like "root cause this", "why is this failing", "debug systematically", "this test is flaky", "it works locally but not in CI", or when a fix attempt has already failed once. Enforces a phased evidence-first process before any code…
canvas-webapp
Render a React/Vite (or any bundled) web app on the pi-dashboard canvas, which loads loopback URLs in a sandboxed opaque-origin iframe. Use when a canvas(target:{kind:"url"|"server"}) target shows up blank white, an empty surface, or a /live/ 500 ECONNREFUSED. Covers why Vite dev servers and non-CORS static servers…