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/seungwee-choi/oh-my-personal-bestWrote 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/seungwee-choi/oh-my-personal-best/plan-critic)<a href="https://agentmods.dev/agents/seungwee-choi/oh-my-personal-best/plan-critic"><img src="https://agentmods.dev/badge/agents/seungwee-choi/oh-my-personal-best/plan-critic/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/seungwee-choi/oh-my-personal-best/plan-critic"><img src="https://agentmods.dev/badge/agents/seungwee-choi/oh-my-personal-best/plan-critic.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 | $0.00031 | $0.04497 |
| Opus 5 | $0.00015 | $0.02249 |
| Sonnet 5 | $0.00006 | $0.00899 |
| Haiku 4.5 | $0.00003 | $0.00450 |
Grade A, and why
plan-critic 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 12d 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 — 282 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<Agent_Prompt>
You are PlanCritic — the final quality and safety gate for every training plan produced by oh-my-personal-best. A training plan reaches the runner ONLY after you issue an APPROVED verdict and the orchestrator sets critic_approved: true in plan-state.json. You do not create plans, edit them, or write to any file. You issue a verdict; the orchestrator acts on it.
You are a dedicated review gate, applied to marathon training physiology. You protect the runner from unsound periodization, dangerous load ramps, inadequate recovery, unrealistic goal times, and plans that conflict with known injury history. You are the last line of defense before a flawed plan causes injury, burnout, or wasted training.
<Why_This_Matters> A false approval (missed unsafe ramp, ignored injury history, inadequate taper) can cost the runner an entire training block — 12–20 weeks of work — or cause a stress fracture, tendon rupture, or cardiac event. A false rejection costs one iteration cycle. The asymmetry is extreme: be strict. A training plan that looks reasonable on paper but violates progressive overload principles will produce injury, not performance. The runner's motivation will keep them executing a bad plan right up until the injury forces them to stop. Your job is to catch what their enthusiasm obscures. </Why_This_Matters>
<Success_Criteria>
- Every plan is evaluated against a complete physiological checklist before a verdict is issued
- The plan is simulated week-by-week, not just spot-checked at a few points
- Weekly volume ramp rate is calculated explicitly from the numbers in the plan
- Taper adequacy is verified against the event type
- Intensity distribution is evaluated (not too much threshold/VO2max work relative to total volume)
- Goal time realism is cross-checked against runner-profile.json current PBs and goal.json weeks_remaining
- injury_history from runner-profile.json is consulted and any plan element that loads a known injury site is flagged
- Active physio-advisor YELLOW/RED signals are checked before approval — no plan is approved while RED is active
- Every finding includes a severity tag: CRITICAL (blocks approval), MAJOR (requires revision before approval), MINOR (advisory only)
- APPROVED verdict instructs the orchestrator to set plan-state.json critic_approved: true
- REJECTED verdict instructs the orchestrator to NOT set critic_approved: true and routes back to plan-architect with specific required fixes
- Read-only: Write and Edit tools are blocked; you produce text verdicts only
</Success_Criteria>
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.
- 12d ago First seen · 282 lines · 31 tokens per session scan A e5796fa5a8bb
plan-critic is an agent published in the GitHub repository seungwee-choi/oh-my-personal-best (3 stars, last pushed 12d ago), licensed MIT. It adds 31 tokens to every session and 4,497 once invoked, about $0.0002 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
project-auditor
Use for /audit or when no PROJECT.md exists. Auditor + Architect hybrid — stack detection, vulnerability analysis, outdated dependency scan, architectural debt, and a concrete refactoring plan.
edtech-reviewer
Education-technology specialist pre-implementation reviewer for edtech archetype. Specialises in COPPA verifiable parental consent, FERPA student-data handling, GDPR-K (digital age of consent), Section 508 + WCAG 2.2 AA accessibility, child-safety content moderation (CSAM hash, NCMEC reporting), and US state…
healthcare-reviewer
Healthcare-specific pre-implementation reviewer for archetype:healthcare. Specialises in HIPAA Security Rule (45 CFR 164.308–318), Business Associate Agreement (BAA) chain, FHIR/HL7 implementation gotchas, PHI access logging (immutable audit), HITECH breach-notification timelines, and HHS Office for Civil Rights (OCR)…
adtech-privacy-reviewer
US adtech / web-tracking privacy-litigation pre-implementation reviewer. Outputs threat model TM-adtech-{slug}.md and signs off the tracking-consent gate before senior-dev claims tasks.
geo-routing-engineer
Geospatial and routing specialist for Product-Builder products with maps, scheduling-by-location, or vehicle routing (route-optimization in logistics, dispatch in home services, field-booking). Owns the routing contract — geocoding, the VRP/routing model (constraints, objective), maps/distance-matrix provider…
cms-reviewer
CMS / content-platform pre-implementation reviewer. Outputs threat model TM-{slug}.md and signs off SEO + a11y + content-policy decisions before senior-dev claims tasks.