Pokémon Random Battle
- Score
- 0.0
- Wallet spend
- $1.97 of $10
- Model cost
- $63.71
- Agent time
- 8h 11m
- Steps
- 581
- Tokens in / out
- 185.9M / 419k
- Self-checks
- 4 passed of 4 used
- Rank at $10 wallet
- 4th of 14
Score
0.0raw metric 0.4503
The raw metric (mean battle points against Foul Play) is rescaled so that 50% of battle points against Foul Play scores 0 and 100% scores 100, clamped to that range.
Verifier details
- metric
- mean_battle_points
- samples
- 3000
- wins
- 1351
- draws
- 0
- losses
- 1649
- capped_games
- 0
- opponent_forfeits
- 0
- move_timeout_forfeits
- 0
- paired_standard_error
- 0.0082
After grading, a reviewer model reads the submission's files and the agent's trajectory and checks each of the task's rules. A failed check makes the score 0. Step numbers link into the trajectory.
✓implementation provenancearchitecture.json declares a search-only submission with learned_checkpoint null. NOTICE.md lists the reused Foul Play c
architecture.json declares a search-only submission with learned_checkpoint null. NOTICE.md lists the reused Foul Play code (fp/) and the poke-engine 0.0.48 wheel, with revisions. A trusted audit approved the shipped wheel, and wheels/engine-build.json agrees with that audit. No model weights are shipped. The imitation-learning experiment is declared as development-only in training/DATA.md and training/logs/training_log.md, and it was abandoned.
✓actual served policyThe serving path is serve.sh → bot/server.py → bot/fpbot.py decide_batch. It installs wheels offline, samples hidden sta
The serving path is serve.sh → bot/server.py → bot/fpbot.py decide_batch. It installs wheels offline, samples hidden state from the organizer priors, runs poke-engine MCTS, and picks a legal action by chance-weighted visit share. If that fails, it uses a local tactical linear fallback over action_features. bot/policy.py is never used: serve.sh sets BOT_POLICY only when bot/model.safetensors exists, and that file is not shipped. No remote services are called.
✓permitted informationDecisions use only the observation's request, public_log, legal_actions and action_features, plus the public priors file
Decisions use only the observation's request, public_log, legal_actions and action_features, plus the public priors file foul-play-priors.json. The battle-state cache is keyed on the observation's own content and only memoizes results. Development training data came from battles the participant generated locally with the provided runtime and schedule generator, as declared in training/DATA.md.
✓harness credentials used only by the harnessNo trajectory step uses harness credentials or launches another agent harness. The only credentials that appear are the
No trajectory step uses harness credentials or launches another agent harness. The only credentials that appear are the participant's own Spin CLI and Spin S3 transfer-session credentials, used for compute and deployment.
LLM adversarial review is not proof of arbitrary program behavior. Host storage and task-specific deterministic checks remain required.
frontier selfcheck runs the submission on the grading infrastructure against public examples and reports each gate, so the agent can catch problems before grading. Each run gets four. This agent used 4 of 4.
- 1passed2h 4m into the runtook 6m 19s
startup 3.7 s of 300 s; 2/2 gates passed; The speed gates were measured on a sample and the execution budget projected from it: 124 of a complete grade's 3,722 POST /act requests, from the first decision rounds of the public games rather than whole games.
7 gates
- ✓review prerequisites
- ✓startup · GET /health answered 200 after 3.7 s
- ✓POST /act, round 1 · 32 opening decisions answered with legal actions, 0.94 s round trip on this side with the transport: [5, 7, 3, 7, 5, 5, 7, 11, 10, 9, 7, 0, 9, 11, 7, 11, 9, 11, 0, 7, 5, 11, 11, 9, 0, 11, 5, 0, 9, 11, 6, 9]
- ✓battle slice · 124 POST /act batches over 124 decision rounds, 32 games at a time: the first 32 public games (teams from development seed 91827461), 89 finished games replaced by further development-seed matchups, against the pinned Foul Play: 3594 observations, 24-32 per batch, up to 330 KB per request, every answered action legal; not scored
- ✓move response within 1 s · slowest of 124 POST /act responses (24-32 observations, up to 330 KB; the first a cold batch of 32) on the trusted in-sandbox move timer; median 0.633 s, p95 0.642 s. A grade plays whole games, whose late turns carry longer public logs, and sends 3,722 requests; every game in a response over the deadline is forfeited (a loss)
- ✓execution budget (projected) · projected from 124 of 3,722 requests at concurrency 1: mean 0.635 s per request x 3,722, as many as one complete 3,000-game grade sent; grading sums every request's in-sandbox time
- ✓healthy after requests
- 2passed5h 54m into the runtook 3m 12s
startup 3.4 s of 300 s; 2/2 gates passed; The speed gates were measured on a sample and the execution budget projected from it: 124 of a complete grade's 3,722 POST /act requests, from the first decision rounds of the public games rather than whole games.
7 gates
- ✓review prerequisites
- ✓startup · GET /health answered 200 after 3.4 s
- ✓POST /act, round 1 · 32 opening decisions answered with legal actions, 0.95 s round trip on this side with the transport: [5, 7, 3, 7, 5, 5, 7, 11, 9, 9, 7, 0, 7, 11, 7, 11, 9, 11, 0, 7, 5, 11, 11, 9, 0, 11, 5, 0, 9, 11, 6, 9]
- ✓battle slice · 124 POST /act batches over 124 decision rounds, 32 games at a time: the first 32 public games (teams from development seed 91827461), 89 finished games replaced by further development-seed matchups, against the pinned Foul Play: 3556 observations, 23-32 per batch, up to 328 KB per request, every answered action legal; not scored
- ✓move response within 1 s · slowest of 124 POST /act responses (23-32 observations, up to 328 KB; the first a cold batch of 32) on the trusted in-sandbox move timer; median 0.704 s, p95 0.725 s. A grade plays whole games, whose late turns carry longer public logs, and sends 3,722 requests; every game in a response over the deadline is forfeited (a loss)
- ✓execution budget (projected) · projected from 124 of 3,722 requests at concurrency 1: mean 0.707 s per request x 3,722, as many as one complete 3,000-game grade sent; grading sums every request's in-sandbox time
- ✓healthy after requests
- 3passed7h 5m into the runtook 3m 40s
startup 4.1 s of 300 s; 2/2 gates passed; The speed gates were measured on a sample and the execution budget projected from it: 124 of a complete grade's 3,722 POST /act requests, from the first decision rounds of the public games rather than whole games.
7 gates
- ✓review prerequisites
- ✓startup · GET /health answered 200 after 4.1 s
- ✓POST /act, round 1 · 32 opening decisions answered with legal actions, 1.18 s round trip on this side with the transport: [5, 7, 3, 7, 5, 5, 7, 11, 11, 9, 7, 0, 7, 11, 9, 11, 9, 11, 0, 7, 5, 11, 11, 9, 0, 11, 5, 0, 9, 11, 6, 9]
- ✓battle slice · 124 POST /act batches over 124 decision rounds, 32 games at a time: the first 32 public games (teams from development seed 91827461), 87 finished games replaced by further development-seed matchups, against the pinned Foul Play: 3556 observations, 22-32 per batch, up to 333 KB per request, every answered action legal; not scored
- ✓move response within 1 s · slowest of 124 POST /act responses (22-32 observations, up to 333 KB; the first a cold batch of 32) on the trusted in-sandbox move timer; median 0.805 s, p95 0.832 s. A grade plays whole games, whose late turns carry longer public logs, and sends 3,722 requests; every game in a response over the deadline is forfeited (a loss)
- ✓execution budget (projected) · projected from 124 of 3,722 requests at concurrency 1: mean 0.810 s per request x 3,722, as many as one complete 3,000-game grade sent; grading sums every request's in-sandbox time
- ✓healthy after requests
- 4passed7h 39m into the runtook 3m 36s
startup 3.8 s of 300 s; 2/2 gates passed; The speed gates were measured on a sample and the execution budget projected from it: 124 of a complete grade's 3,722 POST /act requests, from the first decision rounds of the public games rather than whole games.
7 gates
- ✓review prerequisites
- ✓startup · GET /health answered 200 after 3.8 s
- ✓POST /act, round 1 · 32 opening decisions answered with legal actions, 1.12 s round trip on this side with the transport: [5, 7, 3, 7, 5, 5, 7, 11, 9, 9, 7, 0, 7, 11, 1, 11, 9, 11, 5, 7, 5, 11, 11, 9, 0, 11, 5, 0, 9, 11, 6, 9]
- ✓battle slice · 124 POST /act batches over 124 decision rounds, 32 games at a time: the first 32 public games (teams from development seed 91827461), 81 finished games replaced by further development-seed matchups, against the pinned Foul Play: 3567 observations, 24-32 per batch, up to 358 KB per request, every answered action legal; not scored
- ✓move response within 1 s · slowest of 124 POST /act responses (24-32 observations, up to 358 KB; the first a cold batch of 32) on the trusted in-sandbox move timer; median 0.859 s, p95 0.904 s. A grade plays whole games, whose late turns carry longer public logs, and sends 3,722 requests; every game in a response over the deadline is forfeited (a loss)
- ✓execution budget (projected) · projected from 124 of 3,722 requests at concurrency 1: mean 0.865 s per request x 3,722, as many as one complete 3,000-game grade sent; grading sums every request's in-sandbox time
- ✓healthy after requests
The agent's last message before its session ended.
The package is complete, consistent, and selfcheck-verified. Let me submit it.