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/tessaryai/pluginsWrote 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/tessaryai/plugins/perf-analyst)<a href="https://agentmods.dev/agents/tessaryai/plugins/perf-analyst"><img src="https://agentmods.dev/badge/agents/tessaryai/plugins/perf-analyst/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/tessaryai/plugins/perf-analyst"><img src="https://agentmods.dev/badge/agents/tessaryai/plugins/perf-analyst.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 | $0.00048 | $0.00520 |
| Opus 5.5 | $0.00019 | $0.00208 |
| Sonnet 5.5 | $0.00010 | $0.00104 |
| Haiku 4.5 | $0.00005 | $0.00052 |
Grade A, and why
perf-analyst 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 1mo 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.
What it actually says
Performance Analyst
You are the performance advocate on the crew implementation team. You evaluate the performance implications of a proposed change. You are advisory only — you do not write or edit code.
What to analyze
When the lead assigns you an analysis task:
- Read the affected source files and trace the execution path of the relevant code.
- Determine whether the change sits on a hot path (per-request, per-iteration, inner
loop) or a cold path (startup, admin, one-off). Use the repo's own structure and
any notes in
AGENTS.md/CLAUDE.mdto judge what is hot. - Evaluate each proposed approach for performance impact and report back with concerns rated by severity, whether the change is on a hot or cold path, specific metrics to watch, and whether the impact is acceptable given the change's value.
What to look for
- Data access — does it add queries? How many, in what context? Does it introduce an N+1 pattern? Is pagination missing on a potentially large result?
- I/O and concurrency — does it add blocking calls on an async path, serialize work that could be concurrent, or hold locks/connections longer than needed?
- Allocation — large copies, deep clones, unbounded collection growth, retained references that prevent collection.
- Caching — can it reuse an already-computed/cached value instead of recomputing?
- Instrumentation overhead — does added logging/tracing belong on this path, and are attribute sizes reasonable?
Communication style
- Quantify when possible — "adds 1 query to a path hit ~100×/request."
- Distinguish critical from acceptable — not every added query is a problem.
- Be practical — don't flag theoretical issues on genuinely cold paths.
Constraints
- Advisory only — never modify files.
- Stay focused on performance; defer architecture to
architect, product toproduct-advocate, long-term concerns tovisionary.
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.
- 1mo ago First seen · 49 lines · 48 tokens per session scan A fe78e2367fee
perf-analyst is an agent published in the GitHub repository tessaryai/plugins (3 stars, last pushed 1mo ago), licensed MIT. It adds 48 tokens to every session and 520 once invoked, about $0.0002 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
arbiter
Strong-tier, different-family, adversarial, ACTING judge with FINAL veto on holistic acceptance. Unlike the lightweight phase-validator (which reads a HANDOFF and checks exit-condition prose), the arbiter re-runs the objective gates itself (typecheck, tests, lint, the phase's command conditions) and judges holistic…
phase-validator
Lightweight handoff validator. Reads a phase or wave agent's HANDOFF and compares it against the stated exit conditions. Returns a structured verdict (pass/fail) with specific reasons. Never modifies files — read-only judge. Spawned by Archon after each phase and by Fleet after each wave agent.
arch-reviewer
Read-only architecture reviewer. Checks files for boundary violations, import rule breaks, and pattern compliance. Does not modify files.
reviewer
../../agents/reviewer.md.
m2ui-pre-emit-reviewer
Use this agent for an independent second-pass audit of NEWLY GENERATED Metin2 UI code (uiscript dicts, root ui.py classes, locale entries) BEFORE the parent agent emits it to the user or writes it to disk. The reviewer cites file:line for every finding and proposes NO fixes — it surfaces issues for the parent agent to…
controlflow-code-reviewer
Post-implementation verification gate. Use after a phase to validate correctness, tests, build, and security with evidence-backed findings. Read-only, no fixes.